Dar Ps Manuscript

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/255780364
Doing Archival Research in Political Science
Book · January 2012
CITATIONS READS
8 1,084
4 authors, including:
Scott A. Frisch Sean Q Kelly

California State University, Channel Islands California State University, Channel Islands
26 PUBLICATIONS 209 CITATIONS 36 PUBLICATIONS 320 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Doing Archival Research View project
Jimmy Carter and the Water Wars View project
All content following this page was uploaded by Sean Q Kelly on 03 July 2019.
The user has requested enhancement of the downloaded file.

Doing�Archival�Research
in�Political�Science
EDITED BY
Scott A. Frisch, Douglas B. Harris,

Sean Q Kelly, and David C.W. Parker
Copyright 2012 Cambria Press
All rights reserved

Printed in the United States of America
No part of this publication may be reproduced, stored in or introduced

into a retrieval system, or transmitted, in any form, or by any means
(electronic, mechanical, photocopying, recording, or otherwise),
without the prior permission of the publisher.
Requests for permission should be directed to:

permissions@cambriapress.com, or mailed to:
Cambria Press
20 Northpointe Parkway, Suite 188
Amherst, NY 14228
Library of Congress Cataloging-in-Publication Data
Doing archival research in political

science / Scott A. Frisch ... [et al.], editors
p. cm. . -- (Politics, institutions, and public policy in America
Includes bibliographical references and index
ISBN 978-1-60497-802-5 (alk. paper)
1. Political science--Research--Methodology. 2. Political
science--Archival resources. I. Frisch, Scott A., 1964-
JA86.D65 2012
320.072--dc23
2012003352
Table�of�Contents
List of Figures ................................................................ vii

List of Tables .................................................................. ix
Introduction: Taking the Road Less Traveled
Scott A. Frisch, Douglas B. Harris, Sean Q Kelly, and
David C. W. Parker .................................................. 1
Part I: Political Science and the Archives .............................. 33
Chapter 1: Political Science and Archival Research
Scott A. Frisch and Sean Q Kelly ............................. 35
Chapter 2: Behavioral Reality and Institutional Change
Douglas B. Harris .............................................. 61
Chapter 3: Pulling Back the Curtain
Linda A. Whitaker and Michael Lotstein .................... 101
Part II: Political Science in the Archives .............................. 159
Chapter 4: Why Archives?
David C. W. Parker ........................................... 161
Chapter 5: If This Is Tuesday, It Must Be Albuquerque
David C. W. Parker ........................................... 169
Chapter 6: The Search for the Elusive Executive
Brandon Rottinghaus .......................................... 203
Chapter 7: Rummaging Through the Attics of the CIA and
Congress
David M. Barrett .............................................. 227
Chapter 8: Public Opinion in the Archives
Amy Fried ...................................................... 243
vi Doing�Archival�Research�in�Political�Science
Chapter 9: Digging Through Trash

Kathleen Sullivan and Patricia Strach ...................... 275
Chapter 10: In Search of Influence
Tracy Roof .................................................... 299
Part III: On the Road .................................................... 329
Chapter 11: Hitting the Road Without Hitting the Potholes
Sean Q Kelly and Linda A. Whitaker ...................... 331
Appendices .................................................................. 355
Index ......................................................................... 359
About the Contributors ..................................................... 371
List�of�Figures
Figure 1: Page 1 of a list of members that the Carter White

House liaison team suggested President Carter contact.
The handwritten notes are the president’s, reflecting his
thoughts on each call. ........................................... 44
Figure 2: Whip counts from the papers of Representative Tom

Bevill help illustrate how he built a coalition in support
of his Tennessee-Tombigbee Waterway project ............... 45
Figure 3: Archivist Mary Anne Hamblen in the storage warehouse

with the Ted Stevens Papers Collection (2010). ............. 120
Figure 4: Personal and political papers of Senator Barry M.

Goldwater before processing (2003) .......................... 123
Figure 5: Personal and political papers of Senator Barry M.

Goldwater after processing (2009) ............................ 124
Figure 6: Payroll, Department of Public Works, 1883 ................. 288

List�of�Tables
Table 1: The data available in archival collections. ...................... 49
Table 2: Ready and latent data in congressional papers

collections. ......................................................... 50
Table 3: Container list from the 1964 presidential campaign series

in the Goldwater papers. ........................................ 141
Table 4: Senator John Glenn and going home, 1987–1992. ........... 180
Table 5: Senator Barry Goldwater and going home, 1975–

1980 ............................................................... 180
Table 6: Comparison of Republican vote percentages in

Republican, swing, and Democratic precincts in Maricopa
and Pima Counties, 1976–1980 ................................ 183
Table 7: Change in voter turnout between 1976 and 1980 in

Republican, swing, and Democratic precincts in Maricopa
and Pima Counties ............................................... 184
Table 8: Library of Congress presidential manuscript

collections ......................................................... 206
Table 9: NARA presidential library websites and locations ........... 207
Table 10: Budget worksheet ............................................... 355
Table 11: Travel checklist ................................................. 357

Introduction
Taking�the�Road
Less�Traveled
Scott A. Frisch, Douglas B. Harris,

Sean Q Kelly, and David C. W. Parker
In the discipline of history, archival research is common. In political

science, it is not. Students majoring in history, and certainly history
graduate students, are exposed to the practice, and academic historians
routinely use archival sources in their research. This is not the case for
political science students or scholars. In fact, in political science, the term
archive is most often preceded by the word data, indicating an electronic
repository of (usually) quantitative data.
This book is aimed at political scientists. We simultaneously intend to

advocate for as well as to illustrate and instruct the “how to” of archival
research. In particular, we hope to reach graduate students in political
science who, lacking the pressures of the tenure track, have a unique
2 Doing�Archival�Research�in�Political�Science
opportunity to explore different—and, we argue, more innovative and

fruitful—approaches to research. But we also hope to reach intrepid junior
faculty and even more senior political scientists who are interested in
expanding their approach to the study of politics and traversing a road less
traveled.
In conceiving this book, we hope to produce a volume that we wish had

existed when we first started working with archival collections. All of us
were “classically” trained political scientists, which is to say that we had
no training in archival research and rarely heard archives mentioned in
political science circles. We each found our way into archival research, but
we each faced a steep learning curve and little to no direction written from
the perspective of political science. We also encountered mostly skep-
tical colleagues who were hardly models of encouragement. We hope this
book will be helpful to scholars before they hit that steep learning curve,
offering the necessary encouragement to them to follow a road that few
political scientists, at this point, have dared to travel.
What�Are�Archives?
Broadly speaking, archives are collections of records—both paper and
electronic records—that are generated by, and reflect the efforts of,
an individual, organization, or institution. The archival records that we
discuss in this volume have been generated by politicians, political orga-
nizations, and political institutions.
Archives are more than the boxes of papers one keeps in one’s garage,
however. They are records that have been judged, usually by an archivist,
as being worthy of preservation; they are considered to contain informa-
tion that may be important for understanding some aspect of the human
experience. In short, someone has decided that the records are important
enough to be stored in controlled conditions and that the public should
have access to them.
Introduction 3
Once the decision is made to preserve a collection, the next step (usually
before providing access) is to describe the collection. It is the role of the
archivist to create order out of chaos—that is, to process the collection.
An archivist needs to survey the collection to gain an understanding of the
scope of the collection (e.g., what years it covers and what topics are repre-
sented in the records). From a researcher’s point of view, this is important
for determining whether material in the collection might be useful for a
research project. Usually an archivist will also summarize the contents
and report the provenance of the collection, or how it was generated and
how it arrived in the repository.
Collections are rarely in an order that makes them easy for researchers
to use. Boxes and files are not neatly arranged for easy access. An archivist
is responsible for processing a collection. This involves (to a greater or
lesser extent) putting boxes and files together into topical order—that
is, arranging like materials into groups or series. The degree to which
archivists impose order—often chronological—on a collection is a matter
of scholarly debate within their field, but imposing some order is the result
of arranging the materials.
Archivists are also responsible for creating a finding aid: a written list of
the records that helps the researcher in locating materials within a collec-
tion of documents. At a minimum, a finding aid will describe the kinds of
documents that are included in each box. Finer-grained finding aids will
list the titles of the folders in each box (usually titles that were on the orig-
inal folders or titles that the archivist assigns to a folder), providing some
insight into the documents within the folder. In some cases (the best case
from the researcher’s perspective), archivists will add further description
of the documents contained in the folder.
Once a collection is processed, it is made available to researchers.

Access to the records is controlled by the archivist, subject to the policies
of the repository and the donor’s restrictions. The archivist is now the vital
link between the researcher and the subject of the research.
Because there are many collections, and many of them are very large,
not all collections are fully processed (from the researcher’s point of
view). Repositories must make choices about whether to process a collec-
tion, which collections to process, and how many resources to dedicate
to a collection. As a result, the quality and detail involved in processing
a collection will vary from collection to collection. However, this can be
partially overcome as the researcher develops more experience working
with collections; one begins to develop a sort of “sixth sense” with regard
to them.
Archival collections are located everywhere: from the Library of
Congress to university libraries to government agencies at federal, state,
and local levels to even private corporations. Some manuscript collections
are held in family or private custody. Finding archives, however, can take
some work because there is no one central depository listing all of the
archival collections. The two most complete lists of archival collections
are Proquest’s Archive Finder, which is a “directory of over 5,750 repos-
itories and over 206,000 collections of primary source material housed
across the United States and the United Kingdom,” and the Northwest
Digital Archives, containing “access to descriptions of primary sources in
the Northwestern United States.”1 Finally, those scholars who are inter-
ested in studying Congress should consult the Congressional Biograph-
ical Directory, which contains information on the research collections
of individuals serving in the House and Senate.2 Users need simply to
input the name of the member and click on the research tab. This will
reveal whether papers for the member exist, where they are located, and
some brief information on the size and scope of each collection. There
are archives, in short, for every breed of researcher containing every kind
of material imaginable. They offer limitless research opportunities that
depend only upon the imagination of the researcher, the time he or she
has to do research, and the size of the research grant he or she may have.
Nevertheless, before embarking on archival research, one should proceed
with eyes wide open. There are challenges and pitfalls to consider, and it
is to those we now turn.
Introduction 5
Reasons�Not�to�Do�Archival�Research
In a book partly aimed at encouraging archival research, it may seem
counterintuitive to provide a section on the reasons to avoid it. We do not
mean to discourage archival research but to promote it by addressing the
common sources of hesitation that many readers may have about under-
taking such work.
Reason One: This May Not Generate Publishable Findings

Archival research is time-consuming. Many hours and even days will be
invested traveling to archives and collecting documents. Many more hours
may be spent putting together a data set or constructing a usable interpre-
tation of events based on the documents. What if, after all of this effort,
one is confronted with a computer screen full of null findings? What if
one does not find anything?
This is a fair concern. But, it is equally likely that if one invests consid-
erable time constructing a data set of similar scope from easily avail-
able, previously published data, one faces the same prospect: having spent
many hours and days constructing a data set, only to be left with a heap of
equally difficult-to-publish findings. Most have heard of—or even experi-
enced themselves––the agony of assembling a data set over a considerable
length of time (perhaps years), cleaning that data set of errors, writing code
for days, only to write a logit routine, press “return,” and presto! Nothing.
Zippo. Nada. Not a single significant coefficient. In any case, quantita-
tive, qualitative, statistical or not, one can still end up with research that
belongs in the venerated Journal of Null Results.
So, we doubt that archival research is any more likely to yield null find-
ings than traditional political science research built on quantitative data
sets. In fact, we boldly argue that one is less likely to have null findings
when using archival data (we will even italicize that statement). There
are two reasons for this. First, research that uses archival data relies on
data that are appropriate for testing political science theories. For instance,
congressional scholars have long used committee assignments as an indi-

cator of members’ preferences, the reasoning being that if a member sat
on a particular committee, that was the committee on which the member
preferred to sit. That is, of course, circular logic, but collecting committee
request data, which give insight into members’ actual preferences, was
considered “too difficult”—until it got done, that is (Frisch and Kelly
2006). Null findings are more likely to result from using inappropriate data
—that is to say, using existing data sets that were built to answer different
questions than those that most scholars would like to test.
Second, doing archival research (as we argue in the following section)
helps with theory building. Direct engagement with the data allows the
researcher to explore alternative explanations for null findings, more accu-
rately specify models, refine existing theories, and build more satisfying
theories. The result is more interesting, satisfying, and “positive,” and thus
easier-to-publish, findings. Anyone who is familiar with Richard Fenno’s
work on Congress—although it is not archival in nature but faces similar
challenges and opportunity costs—will easily agree that “thick descrip-
tion” can yield rich theoretical findings. These findings push the discipline
in new directions by building as well as testing theories and by describing
—in detail—the behavioral reality that is attendant to political institutions
and exhibited in politicians’ actions and choices.
We are not making any guarantees here. We are simply arguing that
the odds, based on logic and experience, favor a more profitable and
successful analysis.
Reason Two: The Information Sought May Not Be Found

It is certainly possible that one could make a trip to an archive and find
little to no material that bears on one’s research question. It is possible.
And that would be bad. But it is fairly preventable. Careful work prior
to the trip aimed at avoiding the problem will prevent one from making
unnecessary trips. The advice offered throughout this book, and especially
in chapter 10, will help researchers to avoid fruitless journeys.
Introduction 7
But it could and has happened. Parker nearly experienced cardiac arrest
on one research trip. He was studying two Senate campaigns—Idaho
senator Frank Church’s defeat in 1980 to Congressman Steve Symms
and Symms’s near-loss in 1986 to Idaho governor John Evans. The 1980
campaign was well documented in both the Church and Symms archives.
The 1986 campaign, however, was not well preserved in the archival
record. In particular, the polls commissioned by the Symms campaign in
1985 and 1986 were nowhere to be found in the Symms papers located at
the College of Idaho in Caldwell. The polls undertaken by Governor Evans
were not located in his papers at the Idaho Historical Society. However, a
subsequent interview with Evans’s campaign manager yielded the name
of Evans’s pollster, who—fortunately—had kept some of the polls from
the 1986 race. Parker received the polling data in the form of a PDF the
following day after calling the polling firm directly. The archives informed
Parker that polls existed—but he had to go through some additional steps
to find them. The lesson is that persistence and creative thinking can over-
come a disappointing archival experience.
Remember a piece of advice that nearly every author in this volume

offers: keep an open mind. Be open to the serendipity involved in archival
research. To paraphrase the Rolling Stones, you may not find what you
want, but you find sometimes that you get what you need. As Frisch and
Kelly recount briefly in chapter 1, their book on Jimmy Carter was the
result of a “failed” trip to the Carter presidential library. By maintaining
a broad awareness of the issues that were covered in files related to their
topic, they discovered the means to test presidential influence in Congress
using new, innovative, and particularly appropriate data.
Reason Three: A Great Deal of Time Is Needed

Archival research requires a time commitment. In our experience, trips
average about three days. That is a nontrivial amount of time away from
work and family. But one should ask oneself: How many hours in the last
year have been spent drinking coffee or having lunch with colleagues?
How many hours have been spent in committee meetings? Think about
all the ways time has been wasted at work over the last year. If time can
be found for these things, surely time can be found to spend in an archive,
with all of the many benefits such an excursion promises, as we describe
in the next section.
In short, researchers have the time; it is just a matter of how they

choose to spend it. Applying Newton’s First Law of Thermodynamics, we
can explain this element of resistance to archival research: an ass at rest
tends to stay at rest. Political scientists are socialized to believe that some
behaviors are more valuable than others. Sitting in front of a computer
screen and downloading data sets from the Interuniversity Consortium for
Political and Social Research—and spending time testing hypotheses with
inappropriate data—is acceptable (thereby keeping the ass at rest); trav-
eling to collect one’s own, more appropriate data is a “waste of time.”
Archival travel, actually, is a fantastically efficient use of a researcher’s

time. Consider the following: A researcher travels to another city in
another state, leaving his or her daily responsibilities behind. He or
she works in a library—there are no cell phones, no e-mail, no pesky
colleagues knocking on the door. The researcher’s full attention is devoted
to research—something that probably has not been possible since graduate
school, especially if one is an amiable, likeable, fun person with a social
life. When the archives are closed, the researcher can spend the “down
hours” tracking down leads uncovered from the research, conducting
interviews, or even working on other projects. Parker has been known
to crank out other, more “traditional” political science research projects
while he is on the road—where he is far away from his DVD collection
and Xbox.
Reason Four: Money Is Needed to Do This Research

Money and time are the two most important resources for doing
archival research. Several sources of funding are available. First is
the archival repository itself. Many special collections departments
have travel stipends available. The Congressional Papers Roundtable
Introduction 9
of the Society of American Archivists keeps a list of these funds at

http://www.archivists.org/saagroups/cpr/grants.asp. It never hurts to ask a
repository if they have available funding. Second, many universities have
small internal grant funds aimed at supporting research. We have all been
successful at finding support this way. Other organizations, such as the
Dirksen Center, also support research on a competitive basis.
The good news is that one does not need a lot of money to do archival
research; we are not talking about National Science Foundation levels of
funding here. A thousand dollars is usually sufficient to fund a trip of a
week’s duration or more if one is careful with one’s budgeting. This may
not involve a stay at the Biltmore Hotel, but Holiday Inn Expresses and
Fairfield Inns are perfectly clean and respectable places to stay. Some of us
have even funded archival travel by simply dipping into our department’s
travel budget. Indeed, the value of archival research to one’s teaching
could open up pedagogical or instructional enhancement resources that
are available at some colleges and universities to fund research trips to
archives. In any event, a supportive department chair or dean can go a
long way to advancing the archival researcher’s conquests.
One of the benefits of having to hunt for research support is that early-
career scholars can begin to build a record of accomplishment. Writing for
grants of support and receiving these relatively small amounts of funding
provides evidence of research success. Although it is not a full-blown
publication, it aids in developing and refining research design, is an indi-
cation of research effort (hopefully successfully funded), helps to develop
grant-seeking skills, and results in a line on the curriculum vitae. During
a period in which a researcher is publishing, grant-seeking helps to fill the
“holes” in his or her vita; it indicates that he or she was moving toward a
larger goal and not simply playing Mine Sweeper™ in the office.
Reason Five: Colleagues Say It Is a Waste of Time

Some political scientists may describe archival research as a waste of time.
First, find out why they think this. Is it the research design or the proposed
use of archival sources? If it is the former, a return to the drawing board

may be necessary. If it is the latter—which is more likely—then consider
the source of the criticism. Have they ever used archival sources? Are they
familiar with any research that uses archival sources? Chances are that
they are projecting their own apprehensions and anxieties. Or, they are
simply jealous that they will not be able to travel to West Branch, Iowa,
for a week at the Herbert Hoover Presidential Library and Museum.
No doubt many colleagues believe that they are offering sage advice; we
do not mean to cast aspersions on their motives. But the truth of the matter
is that they are not speaking from experience, they are speaking from their
casual (nonsystematic) observation of the state of the discipline. In the
prebehavioral era of political science, imagine how traditional scholars
must have viewed their young colleagues who were conducting surveys
and coding congressional roll call votes. More than a few younger scholars
must have been discouraged from this work for the very same reasons
we are discussing here. Back then, it surely was easier to make another
trip over to the library to check out some more books to cite in that new
treatise on Aristotle.
If colleagues have reservations about a research design, listen very care-

fully. If their concern is with the archival approach, listen carefully and
test their depth of knowledge; ask them why using existing data is better
than collecting one’s own, and (most of all) refer them to this book.
Five�Reasons�to�Consider�Taking�the�Road�Less�Traveled
Part of our purpose in putting this volume together is to encourage archival
research and to smooth scholars’ introduction to this mode of research. In
contemporary political science, archival research is uncommon; it repre-
sents a road less traveled by political scientists. As advocates of archival
research, we firmly believe that there are a variety of benefits to taking
this road. Some of those benefits are outlined in the following sections.
Introduction 11
Reason One: It Will Improve Research

The chapters in the volume are intended both to instruct and to demon-
strate that archival research makes a valuable contribution to political
science research. In the spirit of methodological pluralism, we focus on
both qualitative and quantitative examples. There are qualitative data and
quantitative data to exploit in these collections, and new data sources can
naturally improve the quality of one’s research.
Where archival research takes the researcher beyond simply some new
data is in the degree to which the archival research process exposes one
to important aspects of process, context, sequence, and timing that are
important for theory building, data collection, and hypothesis testing. To
be brief (since the chapters illustrate this better than we can here), working
through archival records provides a look “behind the curtain” of political
processes. Most of political science focuses on behaviors, on outcomes,
that are usually the result of a more complex process. Peeking into that
ordinarily opaque process can help to build theories of political behavior
that are more reflective of the actual political process. Although this may
seem to some to be mere description, good theorizing depends on accurate
descriptions of political processes.
Wading through the archival record also makes one more aware of
the context within which politics happen. Decisions are not made in a
vacuum, and decision makers are not singular actors. Researchers often,
for good reasons, treat behaviors as if this were the case. They say things
like “member X voted this way” or “the president decided to do X,” but
these decisions were made in a particular context—with a certain under-
standing of policy choices, political conditions, staff advice, and the like
—that produced a unique outcome. Archival research makes one sensitive
to variables that may be useful in explaining what produced a particular
political outcome. To put it in the terms of quantitative research, knowl-
edge of process and context can improve model specification.
Reason Two: It Will Inform Teaching

Research in the papers of former members of Congress, a presiden-
tial library, or some other collection has informed us as teachers and
scholars. Each trip provides insight into the institution and the individ-
uals who comprise it that goes well beyond our more narrow research
focus. Students often learn best from the tangible, and photocopies of
documents provide illustrations of many concepts relating to the topics
that we teach. For example, while digging (literally) through the as-yet-
unprocessed papers of Representative Bob Livingston (R-LA), Frisch and
Kelly discovered several documents, including ones from Ralph Reed and
Ari Fleischer proposing communications strategies for Livingston, who
was at the time the likely successor to House Speaker Newt Gingrich.
The one-page memo from Fleischer to Livingston provides students with
an excellent example of the ideas behind message control that pervade
contemporary politics. It is one thing to discuss the emphasis on message
control and communications strategy; it is quite another when students
see concrete evidence from individuals who are powerful actors in the
contemporary political process.
The following paragraph from Fleisher’s memo to Livingston is a gem

that conveys a great deal in a few words:
The way to get a message out is to constantly repeat the mantra.

In this case
• Education, locally controlled

• Tax cuts
• A stronger Social Security system
• Economic growth
• Honesty and ethics in government
Literally, this is an eight-second sound bite. I recommend you

repeat it ad naseum [sic]; only when you’re sick of saying it will
anyone have heard it.3
Introduction 13
For years, political scientists have sought to explain how politicians try
to control the agenda through a carefully crafted, concise, and frequently
repeated message. This memo, from a future presidential press secretary
to a man who nearly became Speaker of the House, captures the essence
of political communication in a frank and unguarded memo that can be
shared with students.
Other examples of documents found in congressional papers that can

be used in the classroom abound. Students are often curious about
how members are placed on committees. Copies of letters requesting
committee assignments are great ways to illustrate the multiple moti-
vations that members have for pursuing committee assignments. Staff
memos can be used to illustrate the important role played by staff as well
as the influence of lobbyists and policy analysis in the legislative process,
press releases are a great way to illustrate concepts such as credit claiming
and position taking, and the actual pages of a marked-up bill can be used
when discussing the mark-up process. Polling data from campaigns help
students understand campaign strategy, positioning, and the ability (or not)
of candidates to communicate effectively to voters. Though we can go
on and on, the point is that every trip a researcher makes to an archive
provides additional ideas and resources that can be applied in the class-
room.
Reason Three: New and Better Data Will Be Introduced

As a field, it sometimes seems that political scientists are focused more
on developing new and increasingly sophisticated ways to analyze data
than they are on finding appropriate data to answer key questions. It seems
that article after article is published in the leading journals consisting of
new ways to look at roll call data, campaign spending data, or Amer-
ican National Election Studies data. Casual observation suggests that less
emphasis is being placed on efforts to develop better data sets that are
more specifically designed to answer the research questions that are being
asked.
To take just one example, consider that one of the factors that is
seemingly correlated with favorable impressions of incumbents and their
reelection is favorable attention received by the media. When an incum-
bent fails to get favorable coverage, when coverage is lacking, or when
challengers successfully garner media attention, this spells trouble for the
incumbent and might signal his or her defeat. Often, incumbents blame the
media for slanted coverage or inattention—holding them responsible. But
this raises a question: Is it the media’s fault or the incumbent’s inability or
unwillingness to seek that coverage? The easy way to test an incumbent’s
relationship with the media is to search Lexis-Nexis Academic for all of
the articles about the incumbent in a state’s newspaper (if that newspaper
happens to be in Lexis-Nexis). The hard, but better, way is to analyze the
member’s own efforts to obtain coverage by reviewing the press releases
the member chose to send out. That will help one understand how the
member wanted to be portrayed and can quickly reveal whether the media
did indeed slight or negatively portray the incumbent. Did the member
send out many press releases, only to be ignored by the press, or—more
likely—did the incumbent have a poorly planned media strategy which
simply did not attract much media attention? The data located in archives
frequently provide better information to analyze political questions than
do much of the easy data one can download with a click of the mouse.
Reason Four: Faulty and Selective Memories Are Less Common

Each data collection method has advantages and disadvantages. We advo-
cate using multiple methods as a way to overcome the drawbacks of using
a single methodology. Make no mistake: we all use interviews. Inter-
views are an extremely valuable method of inquiry in political science.
As users of interviews, we have come to see their limitations and to see
archival research as a way to improve upon the data that are gathered
through personal interviews. One problem that we frequently encounter
with interviews, however, is that memories fade. Members of Congress
are extremely busy, and the details of decisions that were made months if
not years before become blurry. In addition, members often have motiva-
Introduction 15
tions to portray their actions in the best possible light, and sitting members
may be unwilling to be completely frank in any interview for fear that
their statements will make their way into the popular media.
The archival record, in contrast, is often not subject to this problem.

Minutes of meetings typically provide a more accurate picture of what
occurred than does a single member’s account relayed long after the
meeting. For instance, Frisch and Kelly interviewed a member of
Congress several years ago about his committee assignments. When he
was asked about his request to be placed on the Budget Committee in
the mid-1980s, he did not recall making that request. The archival record,
however, includes significant evidence that he did in fact request assign-
ment to the Budget Committee. It is possible that he forgot; it is also
possible that he did not want to provide the impression that he did not have
the political “juice” to warrant an assignment to an important committee.
Reason Five: Other Subjects of Research Will Be Uncovered

One of the pure joys of archival research is the unexpected finds and the
materials that spark new research questions and directions. During his first
trip to Tom Foley’s papers, Kelly discovered a large series of whip counts
from Foley’s time as Democratic Whip. Kelly tucked that little find away
as a possible next project (only to discover later that Larry Evans had
targeted those data for his own work!).
Whip counts for leadership races are very common in the collections
of members who sought leadership positions. Harris (2006) and Green
(2007) have made good use of these in their quantitative work on the
determinants of leadership selection, as did the authors of The Austin-
Boston Connection (Champagne, Harris, Riddlesperger, and Nelson
2009), though in a more qualitative and historical examination of intra-
party leadership races. Whip counts are also common in the papers of
members who whipped on behalf of their own personal policy initiatives
(see chapter 1) or committee leaders who sought support for certain posi-
tions in committee.
Senator Tom Eagleton’s papers included a weekly tally of constituent

mail received by the office, broken out by subject. One of the more
intractable questions in political science is how constituent views affect
member behavior. These data allow for a fine-grained analysis of that
very question, and it is very likely that similar data exist in other congres-
sional archives. David Parker, who worked with the Eagleton papers for
his project on Senate campaigns, put a note in his research file to possibly
explore this project in the future. The possibilities are truly endless.
Potential�Weaknesses�in�the�Archival�Approach
Over the years, we have heard a number of potential objections to the use
of archival data. We would be remiss if we did not present these objections
and respond to them.
Reason One: Political Papers Are Sanitized

The most commonly expressed concern is that collections are system-
atically incomplete—that is, that critical information is missing due to
conscious “sanitizing” by donors. Thus, the argument goes, there will be
substantial gaps in the archival record that are created before they arrive at
a repository. Deliberate removal and destruction of documents is particu-
larly problematic because it introduces the statistical problem known as
bias into the collections. Because such sanitation is systematic and not
random by definition, it is especially important for users of archival data
to consider the data generation process and how such bias may affect
attempts at inference. To address the threat of this bias, it is worthwhile to
understand how likely the systematic weeding of papers is generally and
where bias is most likely to intrude.
Archivists work closely with potential donors before papers arrive at

a repository; they negotiate the terms of a gift and, of course, play a crit-
ical role in the transfer of these collections to their repositories. As many
archivists have noted in their correspondence with us, there are logis-
Introduction 17
tical constraints on the ability of a donor to engage in sanitizing records.

These collections are often enormous. It would be difficult for donors to
comb through hundreds and sometimes thousands of boxes looking for
potentially sensitive material in any systematic fashion. Herb Hartsook,
the director of South Carolina Political Collections at the University of
South Carolina, put it this way: “On the whole, our donors do not appear
to sanitize their files. I believe this is because they simply don’t have the
time to wade through hundreds of cartons of material” before getting it to
the repository.4 In the case of politicians who suddenly vacate office or
lose reelection, the challenge of sanitizing records would be compounded
by the real need to move out of an office and ship off records in just a
few weeks. Given that even the best-funded archives are starving for the
resources to painstakingly go through these collections, it is unlikely that
any organization or politician has the wherewithal to accomplish the same
task.
Time and time again, we are surprised (shocked) at the level of frank-
ness that we find in the written record. We also know from our inter-
views that members and staff are often unaware of the depth and sensi-
tivity of the material that is archived. For example, a series of memos to
former senator Dennis DeConcini (D-AZ) from Patty Lynch (legislative
assistant for appropriations) is as frank as it is fascinating in the revela-
tions about bureaucratic infighting, unwarranted requests from other sena-
tors, and other matters concerning the annual treasury, post office, and
general government appropriations bills. The following excerpt captures
the frankness of tone that is common throughout the memos:
General theme: we are concerned that Treasury is not paying suffi-

cient attention to its law enforcement bureaus. As a result, Treasury
bureaus are treated as unwanted “stepchildren” in law enforcement
circles, and Justice, DOD [Department of Defense], and ONDCP
[Office of National Drug Control Policy] run rough shod over the
bureaus…Treasury is never on the Hill, they have no visibility, and
leave lobbying matters to the bureaus…Justice, on the other hand,
is everywhere and FBI has a very active and effective legislative
shop. They show up at our mark-ups and conferences and never

miss an opportunity. They get in the door early and make deals
before Treasury even knows an issue is on the radar screen…If
Treasury wants to be taken seriously, this has got to change.5
This series of memos is chock full of information that would be difficult to

elicit in an interview, but that sheds light on the interactions of Congress
and the agencies of the bureaucracy. For virtually every archive we have
visited, there are similar examples of frank and informative documentary
evidence.
This does not mean that donors are not concerned about potentially
embarrassing materials in the papers. But, in the interest of maintaining
the integrity of the collection, archivists develop strategies to put donors’
minds at ease. Mark Greene, the director of the American Heritage Center
at the University of Wyoming, explained,
In my experience, if there is material in a collection that the member

[of Congress] (or his/her chief of staff) believes is sensitive as far as
the member him/herself is concerned, the inclination is not usually
to sanitize it but to negotiate access restrictions on it—at least, that
is what every repository seeks to have happen in such instances…
members who are particularly sensitive about access to their papers
simply do not donate any of them to a repository, but instead either
keep them in storage or destroy them after leaving office.6
In fact, for anyone who was concerned about embarrassing materials,

disposing of an entire collection would be the wisest (lowest-cost) choice,
rather than investing the massive time and effort that would be involved
in going through a collection page by page. Carl Van Ness, an archivist at
the George Smathers Library at the University of Florida, summed it up
nicely: “I would say that ‘sanitization’ of political collections is more of
a myth than a reality.”7
Going forward, it is increasingly less likely that new collections will

be sanitized for two reasons. First, the advent of computers and e-mail
has tremendously increased the size of collections, making the system-
Introduction 19
atic deletion and removal of documents even more challenging. Recently

retired senator Tom Daschle’s papers, housed at South Dakota State
University, provide a case in point. The physical material alone accounts
for two thousand linear feet—a large collection by any standard. But, in
addition to the physical paper contained in the collection, there is an addi-
tional terabyte of digital data.8 Second, both the Senate and House histor-
ical offices have developed detailed guidelines for record management
and programs of instruction for staffers of new members. The guidelines
and training are designed to train congressional staff in the importance of
proper handling, storage, and maintenance of an office’s records.9 This
type of attention to congressional office record keeping is only likely to
improve the scope and breadth of records that are maintained for future
researchers and, at the very least, mitigate any serious efforts to cleanse
records before their donation and preservation.
Reason Two: Political Papers Do Not Reflect Verbal

Transactions
There is an old saying in politics: “Why write it down if you can say it?
Why say it if you can wink?” The quote is meant to express the feelings of
many political operators that one wants to minimize the “paper trail” when
it comes to political decisions. Some critics of an archival approach argue
that much of the archival record is missing due to the verbal nature of polit-
ical operators. Are face-to-face discussions and telephone calls excluded
from the archival record? In large part they are, but it is not unusual to
find memos that summarize discussions or the occasional transcription
of a phone call. And, when it comes to particularly important requests
—for instance, requests for committee assignments, support for a polit-
ical proposal, and so forth—both the person making the request and the
person receiving it prefer to have it in writing to minimize disputes after
the fact. Thus, much of the verbal culture of politics is reflected in the
written record. The second face of power—the things that happen behind
the scenes and out of public sight—is always hard to observe. Archives
are not the perfect answer, but they come closer than nearly any other form
of data to opening up this world by providing the outsider hints and clues
about the documented parts of that verbal world. With a solid review of
the archival record for glimpses of that verbal world, the researcher will
be better informed for interviews with political participants, who can then
shed additional light on discussions and actions that are not well docu-
mented but only hinted at in the physical archival record. The verbal aspect
of politics is hard to fully capture after the fact, and there is no denying it.
The only real way to capture this fully is through the method of participant
observation, which methodologically has its own challenges, drawbacks,
and opportunity costs.
Reason Three: Political Actors Manipulate the Record

Another criticism of relying on archival research that is occasionally
offered is that staff and members might purposely put documents into the
records that portray themselves in the best possible light. This argument
is reminiscent of a scene from the first Back to the Future movie. Marty
McFly (played by Michael J. Fox) is transported back into the past in his
time-traveling DeLorean. In the scene, he seeks to change the future (thus
ensuring his future existence) by sending a letter to his future self through
the mail.
Setting aside the dubious assumptions of time-travel plots, Marty

McFly had a benefit that the generators of political papers do not: perfect
information about future outcomes. McFly knew exactly which events
were critical to produce a desired future outcome. Members of Congress,
political staff, and the like hardly enjoy the same luxury (if they did,
Congress would do nothing but produce perfect public policy). In the
flow of the congressional workload, and absent the relative omniscience
of Marty McFly, it is difficult for members and staff—as it would be for
anyone—to know which of their actions, decisions, and related documents
will come under the scrutiny of researchers years or decades in the future.
Such machinations would also require knowledge of how researchers use
their records to study topics; most of the politicians and staff that we have
interviewed seemed completely mystified by our interest in their papers
Introduction 21
and were astonished to learn that there was anything of worth in those
boxes.
Are there gaps in the archival record? Absolutely, there are. Upon close
examination, all political science data are somehow incomplete. The ques-
tion is whether the gaps in a researcher’s data—in this case, archival data
—are systematic. If those gaps are systematic, it is a potential problem; if
the gaps are random, they are a nuisance (even a big nuisance), but they
do not constitute a fundamental problem. And, we argue, there are means
by which a researcher can fill these gaps.
Why are there gaps in the archival record? It may be not so much for
nefarious reasons, but for perfectly explainable ones, as Mark Greene
explained:
Gaps in the collections occur primarily because of inconsis-

tent record-keeping (including the destruction of inactive records
taking valuable storage space in the attics of Congressional office
buildings; destroyed simply because the office no longer needed
them and no archival repository had yet contacted the member to
explain the value of some of those records) and deliberate appraisal
by the repository (either carried out prior to transfer by the office,
at the advice of the repository, or carried out at the repository after
transfer).10
There are three likely sources of systematic bias in the archival

records. The first are the decisions that are made at archival repositories
(appraisal). Libraries face difficult choices about which collections and
what material to maintain in the context of limited resources; one of the
main limiting factors is the space to house political collections. According
to Leigh McWhite, a political papers archivist at the University of Missis-
sippi,
Repositories are…making tough decisions about whether they

have the space and resources to store entire congressional collec-
tions…a number of archives are choosing to reject or to destroy
selected series within a collection, like case files or academy
appointments, because of privacy and confidentiality issues and

because of calculations on researcher use versus scarcity of space
and financial resources.11
Archivists, however, do not have a stake in advancing or protecting the

reputation of a member or his or her staff. The decisions they make to
keep or dispense with materials adhere to carefully developed standards of
the profession and are therefore unlikely to create the systematic gaps that
introduce bias which would undermine the development of descriptive
inference.
The second source of bias concerns the political papers within congres-
sional papers. These often contain the personal and campaign papers of
the member and are often produced by the member or his or her campaign
staff. The source of the bias, however, is unlikely to be pernicious. The
world of campaigns is transitory by its very nature, and once a campaign
is over, records are often simply trashed en masse. Campaign staffers
come and go, likely taking some of the campaign records with them. The
records of campaigns are the ones that are most likely to be incomplete
and exhibit gaps, but the nature of the missing record is likely due to
natural processes rather than to some systematic effort to delete particular
records while leaving others behind. That said, the most important docu-
ments—such as polls, targeting memos, and the like—are often distributed
among key campaign staffers as well as to the member’s upper-echelon
congressional staff. It is these memos—which are the most important for
the researcher—that are the most likely to survive the campaign process
intact. Parker has found copies of polls and campaign memos scattered
throughout collections, sometimes in the political papers of a member,
at other times in the papers of congressional staff members who had no
“official” role in the campaign.
The third source of bias is papers missing from the collection outright
because a staff member took them when departing the member’s employ.
Former staffers may decide to retain these records themselves, but they
often donate them separately to the archives. Two ways to avoid this
Introduction 23
problem is to carefully review congressional staff directories, develop a

list of key staff members, search the member’s and other collections for
staff member papers, and—if the staff member is still living—contact the
member directly in order to gain access to these papers.
Although bias can certainly be introduced into congressional collec-

tions, understanding the potential sources of those biases can help mitigate
their effects. Archival collections do not provide a complete account of a
member’s career or a particular political moment or process, but neither do
other sources of data. Paying careful attention to what an archival collec-
tion contains and what it does not pushes an investigator to consider other
sources of information that may be available to fill in the information gaps.
Coping with “Missing Data”

There is another saying in politics: “Don’t let the perfect be the enemy
of the good.” Would it be “perfect” if any and every discussion and deci-
sion in a political setting was documented? Sure. Would it be perfect if
every repository could keep every last shred of paper? Of course. Does
any potential shortcoming in the archival record mean that these data
should be disregarded? No. That would be throwing the baby out with
the bath water. That would make the perfect the enemy of the good.
Archival resources offer, in many cases, a qualitative improvement over
many current measures of political behavior (e.g., committee assignments
as a measure of committee assignment “preference” or press releases as a
measure of a member’s governing style).
Researchers certainly do not hold other measures of political behavior

to such a high standard. Congressional scholars routinely employ roll call
votes to measure individual preferences. Of course they are imperfect.
For instance, many votes are never taken because party leaders prefer to
avoid votes that will cause their members to take positions on controver-
sial issues. How would roll call data be different if these “missing data”
were included? By their nature, roll call votes do not reflect a continuum
of choice on the part of voters. Roll call data would be more perfect
if members could vote on a scale from “strongly agree” to “strongly

disagree.” Some tough votes are conducted using voice votes, yet those
are excluded from the roll call data record. Finally, some votes must be
understood contextually and sequentially. As John Kerry admitted on the
campaign trail in 2004, when he was asked about his vote against spending
for the occupation of Iraq and Afghanistan, he remarked, “I actually did
vote for the $87 billion before I voted against it.”12 Though one may read
Kerry’s remarks as a politician trying to have his cake and eat it, too, they
actually highlight the difficulty of understanding an individual’s position
on an issue by looking at one vote alone. Given all of these shortcomings,
should scholars not use roll call voting data in their research?
Survey research depends on random sampling to measure the attitudes

of a population. On average, random sampling does a good job of gener-
ating data that represent the attitudes and opinions of a much larger popu-
lation. But, it is also possible that a random sample can produce data that
are entirely unrepresentative of the population. Because random sample
surveys can produce misleading data, leading to faulty inferences, one
should not rely only on random sample surveys. However, no serious
social scientist would argue that such surveys should be rejected altogether
because they are occasionally faulty.
All data sources have shortcomings. The question is how the researcher
deals with those shortcomings. Given that there will be gaps of all kinds in
the archival record, how does one deal with the problem of missing data?
Given that politics is, by its nature, a collective enterprise, it is unusual
for written communications to vanish entirely. Researchers can explore
multiple archival collections for evidence. It may be that data missing in
one collection exist in another collection that is closely related.
Conducting interviews is invaluable for the archival researcher.

First, it is an excellent method for deepening one’s understanding of
process. It is also a very good way to establish the completeness and
validity of one’s data collection. For instance, when they are collecting
data regarding congressional appropriations earmarks, Frisch and Kelly
Introduction 25
routinely describe (or even show) documents to interviewees. Interviews

also offer an opportunity to discover whether certain transactions were
routinely documented and whether documentation might be available in
the archival record. They may also prompt modifications in a researcher’s
search strategy; interviewees may suggest looking for documents that the
researcher had not considered as a part of the data collection process.
Summary
Archival research is not without its critics. Many of the criticisms,
however, have little empirical support. Though it is the case that in some
isolated cases, collections are “sanitized” by donors, there is no evidence
that this is a routine process. Because politics is a collective endeavor,
it is entirely possible that data gaps can be filled through searches in
other collections. Although it is true that much of what happens in politics
is verbal, most important decisions do find their way into written docu-
ments. To the extent that this does not happen, researchers may find it
valuable to use interviews to fill in data gaps. In short, though archival
data have potential weaknesses, this is true of all data sources. The ques-
tion is whether these potential problems justify ignoring a rich trove of
potential data. As one archivist put it,
It amazes me that a discipline focused on the realm of politics

largely chooses to ignore the wealth of information available in
congressional collections. Are those political scientists obsessed
with data sets and quantifiable analysis not opting to “limit”
their own research endeavors, sanitizing their own observations
by selecting only easily found indicators that can be numerically
examined?13
Organization�of�the�Book
Broadly speaking, we have three objectives for this book. First, we illus-
trate how archival research “fits” into political science, how it is consis-
tent with current trends in research, and how it can improve contempo-
rary research. Second, we demonstrate how archival sources are used to

provide insight into important research puzzles in political science. Third,
we provide illustrations and examples regarding the “how to” of archival
research. In the beginning, archival research can seem intimidating and
overwhelming; it pushes the bounds of one’s comfort zone. We hope to
expand the comfort zone by demystifying archival research.
The book is organized into three parts. In part 1, three chapters provide
background on doing archival research. The first two chapters in this
part are intended to illustrate how archival research has benefits for both
empirically minded and historically minded research. The third chapter is
intended to provide insight from the point of view of the scholar-practi-
tioners who make collections available to researchers.
Frisch and Kelly argue that archival research is a counterforce and yet
fully consistent with the behavioral turn in political science. V. O. Key and
others argued that political science needed to flee the dusty confines of the
library to observe politics, to turn from excogitation to empiricism. The
value of the behavioral revolution that followed was its focus on observing
politicians and quantifying political behavior. Taken to its extreme, prac-
titioners focused on those behaviors that were most easily observed and
most easily measured; reliance on roll call votes and surveys, for instance,
became the norm. Frisch and Kelly argue for returning to the archives
to, in part, discover data that can improve political science and the under-
standing of politics.
Harris discusses the “historical turn” in political science and how

archival data support the cause of the American political development
(APD) movement. One of the ironies of APD is the degree to which a
historically minded movement has failed to capitalize on such a funda-
mental practice of historical research: archival methods. In some regards,
APD appears to rely on rearranging existing historical accounts that have
been offered by historians to test political science theories rather than
directly engaging the historical record. This may be changing somewhat,
Introduction 27
but one would think that APD would be leading the charge into the
archives; thus far, it has not done so.
Archivists are, at once, academics and practitioners. Archivists undergo
rigorous training at the postgraduate level, grounded in the theory of
their discipline; many of them are active researchers. They also engage in
archival practice—that is, acquiring collections and arranging, describing,
preserving, and making collections available to other academics. The
Academy of Certified Archivists supports a certified archivist program
that ensures broad competency in archival practice. The end product
of archivists’ work is what researchers like us (and hopefully like our
readers) access to promote our research. Whitaker and Lotstein provide
a porthole into the world of archivists. Their chapter demystifies the
archivist’s art for the benefit of researchers; it provides an understanding
of the factors that shape how these collections are accessed and outlines
a common vocabulary for understanding the collections with which
researchers work.
The chapters in part 2 serve two purposes. On the one hand, they illus-
trate how archival materials improved the authors’ research and how using
archives allowed them unique perspectives on important research puzzles.
On the other hand, the authors provide insight into the use of different
kinds of collections. We purposely chose these authors because their work
covers many different (though not all) subfields in American politics:
Congress, the presidency, national security studies, public opinion, public
policy, and interest groups. In so doing, we hope to illustrate the point
that regardless of the topic, archival research can form the foundation of
rigorous political science research.
The concluding chapter in part 3 is aimed at the less-seasoned

researcher. Political scientist Sean Kelly and archivist Linda Whitaker
collaborate—Kelly as a researcher and Whitaker as a scholar-practi-
tioner—to provide advice on how to plan for traveling to and working in
an archive. This advice is intended to help improve the experience so that
it is both smooth and productive.
A�Few�Caveats
This volume has a couple of shortcomings. First, the focus is entirely on
doing archival research in American politics. We do not address using
archives in other countries. We wish we could point the reader who is
interested in that topic to some definitive source in political science, but it
simply does not exist. Second, all four of the editors specialize in congres-
sional studies. For that reason, some readers may find parts of the book
Congress-centered. We hope that the chapters in part 2 offset that tendency
and provide a broad cross-section of examples from other subfields of
American politics. Finally, the volume is not objective; we are advocates
for archival research who believe that almost any research in the field of
American politics will benefit from adding an archival dimension. Despite
the objections raised by some researchers, we believe that archives offer
tremendous insight into a vast array of important political questions. In
short, if one really seeks to understand Congress, the presidency, and so
forth, one must go to the archives; to do anything less is to not fully under-
stand one’s subject matter. Full stop.
Archival�Research:�An�“Aha”�Moment
Sean Q Kelly
My first archival trip was to the Dirksen Center in Pekin, Illinois, in
2000. Scott Frisch had read an article in Roll Call about the newly opened
Bob Michel papers. The story included some intriguing quotes from the
committee request letters in the collection. Scott had gone to Pekin to look
at the material in the Michel papers. Upon his return, he called me and
asked about a possible collaboration. We were colleagues at East Carolina
University for a few years and always hoped to find the right project to
work on together.
We very quickly wrote a grant proposal for the Dirksen Center,

requesting funding to use the Bob Michel papers in Pekin. We were
Introduction 29
awarded a small sum ($750) to cover my travel costs from Niagara Univer-
sity in western New York. My wife, who got her undergraduate degree
from the University of Illinois, had friends in Bloomington-Normal (about
forty-five minutes from Pekin) who agreed to put us up for free and feed
us. So, we packed up our Subaru, put our two-year-old daughter in her car
seat, and drove to Bloomington.
I was not sure what to expect on my first day. It was intimidating.

I was stepping into the unknown; I had never done anything like this.
I drove from Bloomington to Pekin. As usual, I arrived early, so I had
to wait. At the appointed hour, I rang the bell and was greeted by Frank
Mackaman. This was before the Dirksen Center had its own building. The
entire operation was in a space on the second floor of the Pekin Public
Library.
Because Scott had been there a few months before and Frank knew the
collection inside and out, I got directly into the boxes. Frank asked me if
I had ever worked with archival documents before. I was embarrassed to
say “no.” In retrospect, I do not know why I was embarrassed. Political
scientists are not trained to use archives. Why should I know anything
about using them?
Frank explained the basics to me, and I got my first boxes. I took the
first one and set it on the table. Taking out the first folder (committee
assignment requests) and carefully putting a placeholder in the box in its
place, I set the folder on the table and opened it up. Based on the political
science literature, I knew exactly what I would find: letters requesting
assignment to committees explaining how important it was to serve on
the committee because of constituency concerns. As I turned the pages of
the first folder, placed it back in the box, and looked through the next, I
began to realize that the empirical reality of the letters did not comport
with the expectations of the literature. Members were asking to be on all
sorts of committees for all sorts of reasons, only a small number of which
had anything to do with reelection or constituency concerns.
A contrarian and empiricist by nature, I almost immediately under-

stood that the archives were for me. These data would allow Scott and
me to properly test a wide range of hypotheses, addressing a debate
that had largely languished (undeservedly) since the publication of The
Giant Jigsaw Puzzle: Democratic Committee Assignments in the Modern
House by Kenneth A. Shepsle in 1978. Four years later, Scott and I
finished Committee Assignment Politics in the U.S. House of Representa-
tives, which, I believe, is a serious challenge to the accepted wisdom that
members’ committee requests are motivated primarily by constituency
concerns and that the selection process reflects routinized self-selection.
Introduction 31
Endnotes
1. See Proquest’s website at http://www.proquest.com/en-US/catalogs/data-

bases/detail/archives_usa.shtml and the Northwest Digital Archives
website at http://nwda.orbiscascade.org/index.shtml.
2. The biographical directory is located online at http://bioguide.congress.
gov/biosearch/biosearch.asp.
3. Memo from Ari Fleischer to Chairman Livingston and Mark Corallo,
November 9, 1998. Livingston Papers, Tulane University, unprocessed.
Emphasis in the original.
4. E-mail exchange with the authors November 2011.
5. Memorandum from Patty Lynch to Senator DeConcini, November 21,
1991, DeConcini papers, University of Arizona, Box 9.
6. Personal communication with the authors, November 2011.
7. E-mail exchange with the authors, November 2011.
8. See the description of the papers, entitled “Thomas A Daschle Career
Papers,” located at http://www.sdstate.edu/sdsuarchives/daschle/index.
cfm.
9. Karen Dawley Paul, Records Management Handbook for United States
Senators and Their Archival Repositories (Washington, DC: Senate
Historical Office, 2006). See also the House Historians’ website at http://
historian.house.gov/ and Title 44 Chapter 29 of the U.S. Code, which
establishes statute procedures for record management in the federal
government.
10. Personal Communication with authors, November 2011.
11. E-mail exchange with the authors November 2011.
12. Daniel Klaidman and Evan Thomas, “Naysaying Kerry’s Nay,”
Newsweek, March 29, 2004.
13. E-mail exchange with the authors, November 2011.
Bibliography
Champagne, Anthony, Douglas B. Harris, James W. Riddlesperger, Jr., and

Garrison Nelson. 2009. The Austin-Boston Connection: Five Decades of
House Democratic Leadership, 1937–1989. College Station: Texas A&M
Press.
Frisch, Scott A., and Sean Q Kelly. 2006. Committee Assignment Politics in
the US House of Representatives. Norman: University of Oklahoma Press.
Green, Matthew N. 2006. "McCormack Versus Udall: Explaining Intraparty
Challenges to the Speaker of the House." American Politics Research 34:
3–21.
Harris, Douglas B. 2006. "Legislative Parties and Leadership Choice:
Confrontation or Accommodation in the 1989 Gingrich-Madigan Whip
Race." American Politics Research 34: 189–222.
Shepsle, Kenneth A. 1978. The Giant Jigsaw Puzzle: Democratic Committee
Assignments in the Modern House. Chicago: University of Chicago Press.
Part�I
Political�Science
and�the�Archives
Chapter 1
Political�Science�and
Archival�Research
Scott A. Frisch and Sean Q Kelly
Political scientists have always quantified whenever and whatever

they could.
—Theodore J. Lowi (1992, 5)
At least once during their political science methodology coursework,

budding political scientists hear the parable of the man who lost his keys:
Walking down the street one night, one happens upon a man who is obvi-
ously intoxicated and is searching the sidewalk and street for something.
Wanting to be a Good Samaritan, one asks the man what he has lost and if
he would like some help. The man replies “Yes, I dropped my house keys
and I am having problems finding them.” A detective by nature, one asks
the man where he was standing when he dropped his keys. “Over there,”
says the man casually pointing toward the darkness down the street while
looking down and casting his head narrowly back and forth. Curious that
the man is not looking in the area where he indicates that he lost his keys
one asks, “Why are you looking here when you lost your keys over there?”
“Because,” the man answers quite seriously, “it is dark over there, and
there is a street light here so I can see better.” Laughing and shaking their
heads in agreement, the students predictably resolve then and there not
to succumb to the allure of using data that are easy to obtain though they
are inappropriate for the research question. Most proceed to forsake the
resolution, either pursuing a research project because the data are easily
available or using less-than-optimal data to address a research puzzle in
the face of the work involved in collecting the appropriate data. Protecting
themselves from the claim that they are acting like the drunk man in the
story, they subsequently argue that more appropriate data do not exist or
would be too costly to assemble.
In this chapter, we claim that archival research improves one’s ability

to generate appropriate data to better test hypotheses central to the study
of political institutions, individual political behavior, and public policy.1
Archival research is also consistent with the urge of many political scien-
tists to quantify and contribute new empirical observations of political
phenomena. Most importantly, our purpose is to entice political scientists
into using archival collections because it will improve the quality of polit-
ical science research. 2
Before continuing, we should explain the nature of our experience with

political papers and thus some of the limitations of this chapter. Since
2000, we have used more than sixty-five congressional and presidential
papers collections, physically visiting three-quarters of those collections
in seventy-seven separate or combined trips. We have visited small univer-
sities with few congressional collections and extremely limited resources,
and we have visited major congressional research centers, the National
Archives, and presidential libraries. Our quest through this period has been
for appropriate data that allow us to test the major theories of congres-
sional organization (Frisch and Kelly 2006b), theories of presidential
Political Science and Archival Research 37
influence in Congress (Frisch and Kelly 2008), and theories concerning

the generation of pork barrel or distributive spending in Congress (Frisch
and Kelly 2013, forthcoming). Throughout this chapter, we will illustrate
our points by drawing on this research. Given the nature of our experience,
we quite naturally draw on our experience using congressional collections.
Dataheads
Political scientists like to count things. They are dataheads. During the
1930s, 1940s, and 1950s, an important shift was taking place within
political science: the so-called behavioral revolution. The most important
elements of this revolution were an emphasis on methodological individu-
alism and the adoption of the norms of the natural sciences, specifically the
scientific method, in the study of politics. Methodological individualism
holds that the appropriate unit of analysis in the study of politics is the
individual, as opposed to concepts such as classes, parties, elites, and the
like, which were considered epiphenomenal to individual behaviors. The
behavioral revolution sought to adapt some of the methods of the sciences
—or at least political scientists’ understanding of how science “works”—
to the study of politics, especially the empirical tradition, which proceeds
from direct observations of individual behaviors, precise measurements of
behaviors through quantification, and the use of statistical methodologies
to analyze the resulting data.3
The behavioral revolution was—to use the language of psychology

—an act of individuation; it was the point at which political science
declared its independence, once and for all, from history and the human-
ities.4 Central to the behavioral revolution was a focus on data collection
methods distinct from those of the humanities. The legendary political
scientist V. O. Key lamented the slowness of movement toward this new
epistemic orientation:
Professional students of public affairs, both by the limiting circum-

stances of their employment and the habits of work induced by the
tradition and training, have been far too dependent upon the library,
the document, and excogitation. Projects that rely on firsthand
observation and utilize the appropriate techniques for the accumu-
lation of data relevant to the analytic problems deserve priority. All
this is not to deny the utilities of the printed or archival source.
The point is simply that heavy reliance on such materials severely
restricts the range of questions open to investigation. Over the past
quarter of a century social scientists have in varying degree extri-
cated themselves from the toils of the library, but the political scien-
tists have made the least progress in this direction. (1956, 30)
In many respects, the behavioral revolution marked an important and

positive turning point for the discipline. Emphasis on observation and
measurement allowed political scientists to seek out and discover regu-
larities in political behavior and attain a more secure grasp on political
praxis. For instance, the prebehavioral classic in the study of Congress
was Woodrow Wilson’s book Congressional Government, which he wrote
in 1885 as a dissertation while he was at the Johns Hopkins University
in Baltimore. Though he was less than fifty miles from Washington, DC,
and the Congress about which he wrote, Wilson never took the time to
visit and observe the Congress in person. Although it is still a classic,
Wilson’s work is excessively normative and dependent on journalistic
descriptions; it would be difficult to argue that the spirit of observation
that was launched during the behavioral revolution did not have a positive
impact on the study of Congress, at the very least.
By the same token, we believe political science took Key’s advice a little
too enthusiastically. Political scientists abandoned the “toils” of libraries
and archival collections, leaving them to be mined by “pre-scientific”
historians. Meanwhile, political scientists, possessed by the behavioral
spirit, sought out real data that could be analyzed using the latest statis-
tical methods, whose complicated calculations could be completed by
rapidly advancing computer technologies. Reliance on data and statistics
completed the metamorphosis of the study of government into political
science. Important strides were made in understanding the political atti-
tudes and behaviors of voters through nationally representative surveys of

American voters. For instance, in their path-breaking book The American
Voter, Campbell, Converse, Miller, and Stokes (1960) tested the prevailing
wisdom that voters choose candidates based on well-defined ideologies
and found that voters used cognitive shortcuts, primarily party affiliation,
to make their choices at the polls. Through their work, Campbell et al.
established survey research as a dominant data source in American poli-
tics.
Congressional studies qua political science were on the leading edge of
the behavioral revolution. As a subfield of American politics, the devel-
opment of congressional studies provides useful insight into the behavior
of political scientists. Political science tends to be pack oriented; once a
relevant and significant research puzzle is identified, political scientists
are magnetically attracted to it (Arnold 1982). This pack orientation is
magnified when ready data are easily available. One example is the ques-
tion of incumbency advantage. David Mayhew (1974b) first brought to the
attention of congressional scholars the propensity of incumbent members
of Congress to win reelection, usually by large margins, and hypothe-
sized reasons for the increasing electoral security of incumbents (Mayhew
1974a, 1974b). As a result, congressional scholars mined electoral data—
which are neatly and regularly generated in a democracy by the federal and
other levels of government—in search of an explanation for the advantage.
A subsequent finding by Kernell and Jacobson (1981) that established a
link between incumbent campaign spending and the incumbency advan-
tage led congressional scholars to explore campaign finance data (which
were made far more easily available by the campaign spending reforms of
the 1970s and are now made available by the Federal Election Commis-
sion for easy download from the Internet).
Another area intensively studied in congressional politics is the roll call

voting behavior of members of Congress. Studies of the voting behavior
of members of Congress and other legislatures have a long pedigree (e.g.,
Miller and Stokes 1963 and Kingdon 1973). This is true, in part, because
voting on proposals is one of the most important (and obvious) activities

of a legislature. Because legislatures have tended to keep careful records
of votes, these could be easily translated into quantitative data. Reforms
in the late 1960s made recorded votes far more common in Congress,
and today votes are reported and recorded on the Internet almost simulta-
neous to the vote. At one point, political scientist Doug Arnold (1982, 94)
calculated that about 10 percent of the articles on American politics in the
four top journals were on these two topics—the incumbency advantage
in congressional elections and roll call voting in Congress. Imagine, for a
moment, a world in which the congressional reforms did not require teller
voting. A huge component of the literature would not exist.
These examples reveal a second important insight into the behavior

of political scientists: political scientists tend to be lazy or rational (take
your pick). Once a data set has been created and electronically archived,
making it available for broad disciplinary consumption, there is little
incentive for political scientists to pursue the creation of new data. Why
would a rational actor invest his or her resources into collecting new data?
Collecting new data sets is expensive, and it is difficult—except at the
largest universities—to find the resources to support new data collection.
As Arnold pointed out, original data collections are
in the first instance, very expensive, and someone must pay the
bill. Once paid, however, it becomes very inexpensive to dupli-
cate the data files and distribute them to interested scholars. Many
universities pay for the automatic acquisition of such data through
the Inter-University Consortium and then provide free computer
time for analyzing them…From the point of view of the individual
scholar, then, doing research on elections is costless. Complete and
automatic subsidization eliminates the need for either grants or
the investment of personal resources. Small wonder that so many
scholars are doing research [using surveys]…It is free. (1982, 101)
Unlike research which relies on the analysis of existing survey data,

archival research poses legitimate barriers for political scientists. First,
political scientists are not trained in archival research (much less alerted
to the existence of archives), which imposes substantial start-up costs

(locating collections, learning how to use a finding aid, developing
successful information-seeking strategies, etc.) on using these collections.
Second, archival research is expensive: the average cost of our trips has
been about $1,000, including airfare, lodging, and when necessary, photo-
copying. Third, archival research is time-consuming: the modal duration
of our trips is three days, and trips of five days are not uncommon (as
is resistance from our families). In sum, archival research involves many
elements of risk for political scientists. In a discipline that is unfamiliar
with archival data, the prospect of publishing archive-based research is
unclear. In the presence of established research questions and existing
data, why bear the risk?5 As Arnold put it, “The competition is [between]
plenty of free data…that scholars can analyze without ever applying for
a grant, spending a dime, or leaving the comfort of the university” (1982,
103).
Many of the things that interest students of American political insti-

tutions are observable (though we wonder if the observable nature of
things is what makes them interesting). However, many of the things that
are interesting about institutional behavior are not observable. Scholars
lament the lack of understanding of congressional leadership, committee
decision making, presidential influence, and the like. These are all things
that happen behind doors that are not open to researchers. To a large
degree, they are impenetrable, even through the participant-observation
techniques practiced by V. O. Key and Richard Fenno (both of us have
worked in Congress, and we use interviews extensively, having inter-
viewed more than a hundred members, former members, White House and
congressional staff, lobbyists, and journalists). Participant observers very
rarely gain the level of access that is necessary to get inside committee
rooms or the West Wing of the White House, where important decisions
are made. Even Richard Fenno had to deal with closed doors. Participant
observers are “birds of passage”; they tend to have access for short periods
rather than spending years conducting academic research. They pop in
and out for a few days, weeks, or months rather than over the course
of years. Finally, participant observers rarely attain the institutional posi-

tions of influence that are necessary to understanding broad institutional
dynamics. How many political scientists have served as the staff director
of a congressional committee, as a White House legislative liaison to the
House of Representatives, or in another position where they were central
to critical decision-making processes?
As Douglas Harris and David Parker indicate in their respective chap-

ters, archival research overcomes some of the shortcomings of participant
observation. First, it gets the researcher behind the closed doors that are so
hard to penetrate as a participant observer. Documents (and increasingly,
e-mails) give insight into processes that are largely unseen by participant
observers. For instance, a memo from President Carter’s liaison staff and
his handwritten notes provide insight into how he lobbied select members
of Congress on a veto override vote that some scholars consider to be a
turning point in his presidency (see figure 1). Green (2006) and Harris
(2006) used data drawn from archives to pry open the otherwise secre-
tive leadership selection process, an activity that tends to be member to
member and thus difficult to penetrate.
Congressional and presidential collections by their nature document the

entire career of a subject. This means that rather than observing as a bird
of passage, the researcher is able to draw back the curtain on multiple and
long-term decision processes to understand and appreciate the complexity
of individual behaviors. Ultimately, much of that complexity will be
stripped away (rightfully, in many cases) in the interest of identifying
regular patterns, but in the interim, it helps the researcher to be a better
theorist.
Using archival data, we examined the efforts of a House appropriator

(Tom Bevill, D-AL) to protect a major public works project—the creation
of the Tennessee-Tombigbee Waterway—that benefitted his state (Frisch
and Kelly 2006a). Over several years, amendments were offered to kill the
project. Bevill actively whipped House members to support his position.
Collecting the documents allowed us to better understand the process of
member-to-member whipping behavior. It also helped us to better under-

stand the degree to which active whipping can persuade members to
support a powerful member on a critical vote. Bevill ultimately won on a
final, critical vote by encouraging a small number of members to “take a
walk”—to engage in strategic abstention—a behavior that up to this point
has gotten little or no attention in the literature. Of course, why would it?
In the absence of the archival record, there would be no way of knowing
which abstentions were strategic and which were simply from members
who were unavailable to vote.
Figure 1. Page 1 of a list of members that the Carter White House liaison
team suggested President Carter contact. The handwritten notes are the
president’s, reflecting his thoughts on each call.
Figure 2. Whip counts from the papers of Representative Tom Bevill help illus-
trate how he built a coalition in support of his Tennessee-Tombigbee Waterway
project.
Political papers collections contain the records of a member and usually

most or all of his or her staff. A researcher can gain multiple perspec-
tives from within the operation. And, because politics is all about inter-
action, the researcher can gain perspective on behaviors across offices or
across institutions. Our work on Jimmy Carter’s veto of the fiscal year
1979 energy and water appropriations bill used papers from the Carter
administration, but also papers from his chief congressional rival, Tom
Bevill. Both men were feverishly whipping on the critical veto override
bill. Using the archival record, we were able determine which members the
men were whipping, when they were whipping the same member, and the
degree to which they were successful at changing members’ minds. Our
results indicate that direct appeals to members could get them to change
their minds, but when a member was whipped by both men, a presidential
appeal was more effective than a member-to-member appeal. It may seem
an obvious statement when it is read here, but only the data in the archives
have the power to illustrate it and empirically support it.
Oh,�the�Places�You’ll�Go
If one has never used collections, it is hard to imagine their vast potential
and what might be in them. It is all the more difficult for researchers who
have limited experience working in practical politics; one will be surprised
by the degree to which behaviors are documented and retained by offices.
Generally speaking, American politics scholars tend to specialize in polit-
ical institutions, electoral behavior, interest groups, political parties, and
public policy. In each of these subfields, congressional papers have the
potential to yield important data. Table 1 provides a suggestive listing for
the potential user of the papers. What constitutes potential data? A scholar
who is well-grounded in the scholarly literature and carefully studies the
descriptive material that is the subject of his or her study will find that
appropriate materials are easily identifiable:
1. Attitudinal surveys conducted by politicians, political parties, or
political consultants within a political jurisdiction (city, county,
state, congressional district, or state). Nationally representative

samples that allow one to make accurate inferences about national
attitudes do not allow one to understand how local public opinion
impacts the decisions of politicians. Surveys focusing on a specific
jurisdiction can be used to examine the relationship between public
opinion and elite decision making.
2. Whip counts or surveys of other politicians in the same institution,
which indicate support for or opposition to pending legislation or
actions. Though the party leadership often engages in whip counts,
it is not uncommon to find these counts in the collections of rank-
and-file members who are pursuing specific legislative goals or who
are pursuing an elected leadership or other institutional position.
3. Letters from other politicians requesting help or support for a
project or position. These kinds of letters tend to be voluminous in
the case of political actors who served in positions of formal party
or committee leadership. Often these letters will be summarized in
spreadsheets and binders that may have been used in formal deci-
sion-making meetings.
4. Documents that contain information about money but are not typi-
cally made available to the public. Money is the lifeblood of politics.
In electoral politics, money helps to finance elections, and informa-
tion regarding the when, where, and how of campaign contributions
is of great interest to many students of American politics. In public
policy, the battle over money (e.g., how much money will be allo-
cated for particular policies and projects) is critical; unpublished
materials that document proposed and/or actual public expenditures
or campaign expenditures are of great interest to many scholars, too.
One of the pure joys of archival research is the element of serendipity.
In looking for specific documents such as those in the table, a researcher
will run across other, sometimes more interesting and revealing, material
in nearby folders and boxes. For example, our work on Jimmy Carter
began with Kelly traveling to the presidential library to look for evidence
of members of Congress seeking to influence executive branch appropri-
ations decisions. As he worked through the papers, an equally interesting
story and data source presented itself. The congressional liaison staff had
maintained a detailed record of a “veto strategy” and the ensuing battle

with Congress. It became clear that we had an opportunity to examine
how a president sought to influence votes in Congress. In short, the Jimmy
Carter book was a “mistake,” a side trip made possible by serendipity,
which resulted in our meeting and befriending members of the Carter
administration and having a meeting with President Carter himself. In the
words of Dr. Seuss, “Oh, the places you’ll go!”
Another benefit of archival research is the opportunity to visit the places

that produced the politicians who are being studied. Representation in the
American system is largely based on geography; members of Congress
represent states and parts of states, and presidents have often served first
as members of Congress and/or governors of states. Politicians’ beliefs
and behaviors are often shaped by their experiences growing up in the
areas that they represent, the area where their papers are usually deposited.
Visiting these locales can provide insight into how the geography shaped
the politician and how the politician shaped the place that he or she
represented. Much as Richard Fenno urged congressional scholars to
leave Washington to study members in their districts, conducting archival
research gets scholars out to the districts and states of former members
of Congress.
Table 1. The data available in archival collections.

Table 2. Ready and latent data in congressional papers collections.

For more than twenty years, Joseph McDade represented Scranton,

Pennsylvania. Before it was the setting for the fictional television series
The Office, Scranton was home to the coal mining industry and fueled
the Industrial Revolution in the Northeast. As the coal industry went into
decline in northeastern Pennsylvania, it left behind a brutal legacy of envi-
ronmental contamination and mine subsidence that threatened people’s
homes. As a member of the House Appropriations Committee, McDade
made it his business to use the power of the purse to address environmental
problems and promote economic development in the depressed region. As
a result, one drives into Scranton on the Joseph McDade Expressway and
passes by Joseph McDade Park. In Scranton, one can visit the Anthracite
Coal Museum—and take a long elevator ride down into a mine to expe-
rience what it is like to work in a mine—or visit the famous (if not infa-
mous) Steamtown, a train museum operated by the National Park Service.
On the way to Weinberg Memorial Library to work in McDade’s papers,
be sure to keep an eye peeled for the Joseph McDade Center for Literary
and Performing Arts, which is just across the way from the library. “Oh,
the places you’ll go!”
Conclusion
Analysis of quantitative data tends to dominate academic political science.
Beginning in the middle third of the twentieth century, a premium was
placed on measurable and observable behavior. This trend pushed polit-
ical scientists out of the library and increasingly into the computer lab
(and occasionally into the field). The methods of history, archival research
in particular, were to be abandoned once and for all to the lonely and
gnomish historian; political science was prepared to join the ranks of the
more mature social sciences: economics and psychology. Abandoning the
archives was viewed as an indication of progress. Though historical polit-
ical science is experiencing something of a renaissance, an anonymous
political scientist put it this way: “The battle is over, quantitative political
science has won, and everyone gets to learn how to estimate maximum
likelihood models.” Even considering the bravado and overstatement, the

quote illustrates an important point: if archival research is to enter the
mainstream of political science research, it likely will be on the coattails
of quantitative analysis. The majority of articles in top journals are quan-
titative, and they do not rely on archival data (Frisch and Kelly 2009). We
believe that the spread of archival research to political science is depen-
dent on identifying quantifiable data in the preserved documents in the
collections of public figures.
Our formal collaboration began in March 2000. Frisch called Kelly and
told him about some of the wonderful things he had found while he was
working in the Robert Michel papers at the Dirksen Congressional Center
in Pekin, Illinois. Among the wonderful things were hundreds of paper
ballots, ballots that were used by the Republican leadership to vote on
who would be assigned to each of the congressional committees. One
would assume that these ballots would be secret, but “no,” Frisch said,
“the people casting the ballots put their names on the ballots.” Intrigued
by the possibility and funded by a small grant from the Dirksen Center,
Kelly traveled to Pekin, where these not-so-secret ballots had been care-
fully stored in envelopes by some conscientious Republican staffer. Kelly
spent almost three days entering the information contained in those thou-
sands of ballots into a spreadsheet (there were far too many to make copies
on our limited budget). At one point, the archivist and director of the
center, Frank Mackaman, strolled over and said, “I always wondered if
those were of any use.” One chapter of our book Committee Assignment
Politics is dedicated to an analysis of those data, which allowed us, among
other things, to empirically verify factions within the Republican Party at
the time in a way that would have been impossible otherwise.
Political scientists are dataheads, as our experience illustrates; they love

it when they can find things to count and then use them in quantitative
analysis. Though history-minded political scientists continue to work and
publish, large-scale use of archival sources is most likely to come from
the quantitatively minded wing of the profession. Larry Evans’s recent
work (see, for instance, Evans and Renjilian 2004; Evans and Lipinski
2005; Evans et al. 2005; Evans and Grandy 2009) is a hopeful example
of a senior and well-respected congressional scholar embracing archival
research and illustrating both the descriptive benefits and the empirical
power of this approach in the area of congressional studies. If the examples
in this chapter and the other chapters in this volume are not enough to
convince the reader of the benefits of archival research, Evans’s example
should erase all doubts.6
Endnotes
1. For that reason, in the past several years we have applied the zeal
and enthusiasm of evangelists to spread the good news of archival
research. This includes publishing an article in a leading political science
journal (Frisch and Kelly 2003), participating in a joint political scien-
tist–archivist roundtable at a major regional political science conference,
and holding a “short course” for political scientists on doing archival
research at the 2005 American Political Science Association conference.
2. We owe an enormous debt of gratitude to the many archivists and archival
staff that we have had the opportunity to learn from, including Beth
Bower, Rose Diaz, Michael Knies, Rebecca Johnson-Melvin, Richard
Hunt, Jessie Kratz, Frank Mackaman, Charlotte Walters, and Linda
Whitaker, to name just a few. Our work in archives has been expen-
sive, and we have had the financial support of the Dirksen Congressional
Center, the Carl Albert Center, the Gerald Ford Presidential Library, the
Institute for Humane Studies, the Thomas S. Foley Institute at Washington
State University, the Niagara University Research Council, and California
State University Channel Islands.
3. Shapiro (2005) argued that the deductive and empirical elements of the
“behavioral revolution” that we describe are distinct traditions; we present
them as a unified tradition, which we believe is all the more appropriate
given the rise of the empirical implications of theoretical models (EITM)
movement, which seeks to reconcile these two somewhat distinct tradi-
tions.
4. This is not to argue that political science has not maintained a strong
attachment to history, as the American political development movement
illustrates. However, it would be difficult to argue that the top political
science journals are not dominated by behavioral and quantitative polit-
ical science.
5. For junior faculty, under pressure to publish to achieve tenure, the risks
may be even higher. The time-consuming data collection phase of archival
research could substantially slow the rate of publication, raising the
eyebrows of more senior faculty.
6. There are many other examples in the political science literature that
demonstrate the use of archival sources in quantitative data analysis.
Forrest Maltzman, James Spriggs, and Paul Wahlbeck (2000) used the
papers of former Supreme Court justice Harry Blackmun to great effect,

and data from the same source have recently found their way into the
American Political Science Review (Johnson, Wahlbeck, and Spriggs
2006). Harris’s (2005a, 2005b) work on party leadership in the U.S.
House makes extensive use of archival data. Lawrence, Maltzman, and
Wahlbeck (2001) used archival data in their work on committee assign-
ments at the turn of the last century. In addition to our work on House
committee assignments, we used archival data to examine Senate Demo-
cratic committee assignments (Frisch and Kelly 2006b). Terry Sullivan’s
work using presidential papers is seminal (see, for instance, Sullivan 1990,
1991a, 1991b, 2001).
Bibliography
Arnold, R. Douglas. 1982. “Overtilled and Undertilled Fields in American

Politics.” Political Science Quarterly 97: 91–103.
Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E.
Stokes. 1960. The American Voter. New York: Wiley.
Clubb, Jerome M. 1975. “Sources for Political Inquiry II: Quantitative Data.”
In Strategies of Inquiry, Handbook of Political Science, Volume 7, edited
by Fred I. Greenstein and Nelson W. Polsby, 43–77. Reading, MA:
Addison-Wesley.
Evans, C. Lawrence, and Christopher B. Renjilian. 2004. “Cracking the Whip
in the U.S. House: Majority Dominance or Party Balancing?” Presented
at the annual meeting of the American Political Science Association,
Philadelphia, PA, August.
Evans, C. Lawrence, and Claire E. Grandy. 2009. “The Whip Systems of
Congress.” In Congress Reconsidered, edited by Lawrence C. Dodd and
Bruce I. Oppenheimer, 189–215. Washington, DC: Congressional Quar-
terly Press.
Evans, C. Lawrence, and Daniel Lipinski. 2005. “Holds, Legislation, and the
Senate Parties.” Presented at the Conference on the U.S. Senate, Rother-
mere American Institute, University of Oxford.
Evans, C. Lawrence, Sarah Brown, Keith Devereaux, Kristen Haase, Will
Marlow, and Josh McHenry. 2005. “Tax Cuts, Contras, and Partisan Influ-
ence in the U.S. House.” Presented at the annual meeting of the Midwest
Political Science Association, Chicago, IL, April.
Frisch, Scott A., and Sean Q Kelly. 2003. “Don’t Have the Data? Make Them
Up! Congressional Archives as Untapped Data Sources.” PS: Political
Science & Politics 36: 221–224.
———. 2004. “Political Scientists: Strangers in a Wonderland.” Congres-
sional Papers Roundtable, Newsletter, July, 5–7.
———. 2005. “Five Reasons to Consider Taking the Road Less Trav-
eled.” American Political Science Association Legislative Studies Section,
Extension of Remarks 28: 28–33. http://www.apsanet.org/~lss/Newsletter/

july05/EOR-July-2005.pdf.
———. 2006a. “Building Coalitions and Protecting Pork: The Case of the
Tennessee-Tombigbee Waterway.” Paper presented at the annual meeting
of the Northeastern Political Science Association, Boston, MA.
———. 2006b. Committee Assignment Politics in the U.S. House of Repre-
sentatives. Norman: University of Oklahoma Press.
———. 2006c. “Committee Assignment Politics in the U.S. Senate: Demo-
cratic Leaders and Democratic Committee Assignments, 1953–1994.”
Congress and the Presidency 33: 1–23.
———. 2008. Jimmy Carter and the Water Wars: Presidential Influence and
the Politics of Pork. Amherst, NY: Cambria Press.
———. 2009. “Dataheads: What Archivists Need to Know About Political
Scientists.” In The Political Archives Reader, edited by Glenn R. Gray,
L. Rebecca Johnson Melvin, and Karen Dawley Paul, 401–418. Lanham,
MD: Scarecrow Press.
———. 2011. Cheese Factories on the Moon: Why Earmarks Are Good for
American Democracy. Boulder, CO: Paradigm Publishers.
———. Forthcoming 2013. Pork: The Politics of Congressional Earmarks.
Ann Arbor: University of Michigan Press.
Gray, Glenn R., L. Rebecca Johnson Melvin, and Karen Dawley Paul, eds.
The Political Archives Reader. Lanham, MD: Scarecrow Press.
Green, Matthew N. 2006. “McCormack Versus Udall: Explaining Intraparty
Challenges to the Speaker of the House.” American Politics Research 34:
3–21.
Harris, Douglas B. 2005a. “House Majority Party Leaders’ Uses of Public
Opinion Information.” Congress and the Presidency 32: 133–155.
———. 2005b. “Orchestrating Party Talk: A Party-Based View of One-
Minute Speeches in the House of Representatives.” Legislative Studies
Quarterly 30: 127–141.
———. 2005c. “Recovering History and Discovering Data in the Archives:
An Alternative ‘Mode of Research’ for Congress Scholars.” American
Political Science Association Legislative Studies Section, Extension of
Remarks 28: 1–10. http://www.apsanet.org/~lss/Newsletter/july05/EOR-
July-2005.pdf.
———. 2006. “Legislative Parties and Leadership Choice: Confrontation or

Accommodation in the 1989 Gingrich-Madigan Whip Race.” American
Politics Research 34: 189–222.
Johnson, Timothy R., Paul J. Wahlbeck, and James F. Spriggs II. 2006. “The
Influence of Oral Arguments on the U.S. Supreme Court.” American Polit-
ical Science Review 100: 99–113.
Kelly, Sean Q. 2005. “Using Archival Resources in Legislative Research:
Choosing the Road Less Traveled.” American Political Science Associa-
tion Legislative Studies Section, Extension of Remarks 28: 1–10. http://
www.apsanet.org/~lss/Newsletter/july05/EOR-July-2005.pdf.
Kernell, Samuel, and Gary C. Jacobson. 1981. Strategy and Choice in
Congressional Elections. New Haven, CT: Yale University Press.
Key, V. O. 1956. “Strategies in Research on Public Affairs.” Social Science
Research Council Items 10: 29–32.
Kingdon, John W. 1973. Congressmen’s Voting Decisions. New York: Harper
and Row Publishers.
Lawrence, Eric, Forrest Maltzman, and Paul J. Wahlbeck. 2001. “The Poli-
tics of Speaker Cannon’s Committee Assignments.” American Journal of
Political Science 45: 551–562.
Lowi, Theodore J. 1992. “The State in Political Science: How We Became
What We Study.” American Political Science Review 86: 1–7.
Maltzman, Forrest, James F. Spriggs II, and Paul J. Wahlbeck. 2000. Crafting
Law on the Supreme Court: The Collegial Game. New York: Cambridge
University Press.
Mayhew, David R. 1974a. Congress: The Electoral Connection. New Haven,
CT: Yale University Press.
———. 1974b. “Congressional Elections: The Case of the Vanishing
Marginals.” Polity 6: 295–317.
Miller, Warren E., and Donald E. Stokes. 1963. “Constituency Influence in
Congress.” American Political Science Review 57: 45–56.
Shapiro, Ian. 2005. The Flight from Reality in the Human Sciences. Princeton,
NJ: Princeton University Press.
Skemer, Don C. 1991. “Drifting Disciplines, Enduring Records: Political
Science and the Use of Archives.” American Archivist 54: 356–368.
Sullivan, Terry. 1990. “Bargaining with the President: A Simple Game and
New Evidence.” American Political Science Review 84: 1167–1195.
———. 1991a. “The Bank Account Presidency: A New Measure and
Evidence on the Temporal Path of Presidential Influence.” American
Journal of Political Science 35: 686–723.
———. 1991b. “Explaining Why Presidents Count: Signaling and Informa-
tion.” Journal of Politics 52: 939–962.
———. 2001. “Headcounts, Expectations, and Presidential Coalitions in
Congress.” Journal of Politics 63: 567–589.
Vose, Clement E. 1975. “Sources for Political Inquiry I: Library Reference
Materials and Manuscripts as Data for Political Science.” In Strategies
of Inquiry, Handbook of Political Science, Volume 7, edited by Fred I.
Greenstein and Nelson W. Polsby, 1–38. Reading, MA: Addison-Wesley.
Chapter 2
Behavioral�Reality�and
Institutional�Change
Historical�Methods�for�Political�Science’s
Historical-Institutional�Turn
Douglas B. Harris
Archival research provides evidence and insights into many political

science questions in ways that more typical methods and modes of analysis
cannot.1 I know this from experience, having tried my hand at two
different approaches in my analyses of congressional party leaders’ uses
of mass media strategies of legislative leadership (first a more traditional
behavioral-institutional analysis of this phenomenon and then a more in-
depth historical-institutional and archival examination).2 In writing my
PhD dissertation and an article on the rise of the public Speakership
(Harris 1998), I employed some traditional tools of institutional-behav-
ioral analysis: I examined official House websites and congressional docu-
ments as well as published accounts from secondary press sources and
the scholarly literature; I counted the appearances of Speakers on the
ABC, NBC, and CBS nightly news programs as well as the appearances
of congressional party leaders on Meet the Press and Face the Nation;
and I examined the Congressional Staff Directory to chart the increase
of press secretaries and communications directors in House leadership
offices. That analysis charted the rise of the public Speakership and public
congressional leadership in the 1980s and 1990s, characterizing it as an
institutional departure from the more private style that typified congres-
sional leadership since at least the Speakership of Sam Rayburn in the
mid-twentieth century.
Although that analysis is, I think, significant and its conclusions have
withstood scholarly scrutiny, including my own subsequent research on
the subject, it nevertheless necessarily traced the rise of this public style
quite broadly, leaving some important questions unaddressed. One promi-
nent example of an open question was, When and how exactly did the
public Speakership take hold? This question centered on the role of
Speaker Carl Albert. An important transitional figure in the House gener-
ally, Carl Albert was Speaker from 1971 to 1976, where he presided over
one of the most significant periods of institutional reform and change in
House history. Albert was a transitional and ambivalent figure, too, in
terms of the rise of the public Speakership. In some respects, Albert’s
Speakership evinced an increased willingness to “go public.” He hired
the first press secretary in the Speaker’s staff, J. Roddy Keiser, whom he
replaced with Joe S. Foote, who brought a great deal more press experi-
ence to the role. He established the critically important Information Task
Force designed to bring together up-and-coming Democratic leaders who
were knowledgeable about and focused on media and messaging. And, he
made more (albeit only slightly more) appearances on the nightly news
and on Sunday morning talk shows compared to Speaker John McCor-
mack. However, even during the divided government of the Nixon and
Ford presidencies, Albert’s willingness to employ media strategies and
his visibility paled in comparison to subsequent Speakers. The balance of
this evidence led me to lump Albert’s speakership in more with his prede-
cessors Sam Rayburn and John McCormack than with the public Speak-
Behavioral Reality and Institutional Change 63
erships of his successors Tip O’Neill, Jim Wright, Tom Foley, and Newt
Gingrich (Harris 1998).
As I turned my attention from my dissertation and this article to a
broader book manuscript that could address these questions and many
others, I sought to flesh out the historical record of the rise of the public
Speakership and public congressional leadership more generally across
both the House Democratic and Republican Parties from the 1960s to
the 1990s. How, I wondered, should one interpret Albert’s transitional
and ambivalent role in the development of the public style of legisla-
tive leadership? Did Albert ignore the impact of divided government and
party conflict, the generational change of the membership he led, and the
increased importance of television more generally in favor of the received
wisdom and skills he honed during his quarter century in the House under
Rayburn and McCormack? Or, did he sense those changes and respond?
What, if anything, was the nature of his response?
Having found the historical and press accounts of congressional party

communications wanting (trust me—I looked), I was in search of better
evidence (“data”) and a more complete—a more organizationally and
behaviorally real—account of this critical change in House leadership.
Inspired by other recent works that had used archival sources (Maltese
1994; Kolodny 1998), I was determined to recover the history of the rise
of public congressional leadership myself, beginning in the vast archival
holdings of the Carl Albert Center at the University of Oklahoma.
My research at the Albert Center (my first-ever archival research trip)

helped shed important light on a number of open questions, including
Speaker Carl Albert’s ambivalent place in the rise of the public Speak-
ership.3 Uncovered among the documents were extensive notes and tran-
scripts of the Speaker’s daily press conference, planning documents and
minutes from meetings of the Information Task Force, transcripts of
joint meetings between the House and Senate Democratic leadership,
internal correspondence among congressional party leaders and between
leaders and staff, campaign travel schedules, numerous policy documents,
documents related to the Nixon and Ford administrations (including the

impeachment process and resignation), and thousands of other documents
centering around the life and work of the Congress and Washington.
Although I took a broad sweep through these documents, my focus was on
press files, speeches, legislative leadership, and the intersections thereof.
There was considerable evidence in these documents, indeed, not only of
both sides of Albert’s ambivalent stance in regard to public leadership but
also providing clues as to how to read and interpret that ambivalence.
First, there was ample evidence in the archives that, officially and orga-
nizationally, Albert was stepping up his media presence and communi-
cations strategies more generally. There were notes of party meetings
where press strategy (and the need for such strategies) was discussed
as well as letters to the Speaker from House Democrats exhorting him
to greater media visibility and offering advice and tips on how best to
engage the press. The composition of Albert’s Information Task Force,
the minutes of its meetings, and the reports that were written for it shed a
great deal of light on the state of congressional party communications in
the 1970s. And, among many other artifacts, internal memoranda between
the Speaker and his press secretaries offer glimpses into Albert’s office’s
views of the media and political environment they confronted as well as
their press intentions. Early in 1973, Joe Foote wrote a memorandum to
the Speaker outlining strategies and changes to the format of the Speaker’s
print press conferences and those that included broadcast media (what they
termed “news conferences”) to make these media interactions more news-
worthy. These changes, Foote argued, “would give us a more dynamic and
powerful press posture, and at the same time, allow us more control over
our appearances before the press. It should also effect a more equitable
balance between newspapers and the broadcast media.”4 Though they are
seemingly rudimentary in retrospect, these initial efforts to embed an orga-
nizational media presence in the House leadership are noteworthy, partic-
ularly in light of the fact that prior Speakers had avoided media strategies
of legislative leadership. In some respects, Carl Albert was a pioneer.
Still, archives reveal, too, that Albert’s first behavioral steps toward
media leadership were cautious, reluctant, and prone to mistake, as one
would expect given his House experience and the skill set he accumu-
lated during his long tutelage under Rayburn and McCormack in the mid-
century House. Although a complete consideration of this is beyond the
scope of this chapter, note, for example, how if Albert officially recog-
nized the dawn of congressional media politics, he nevertheless seemed
personally reluctant to “go public.” It is clear in the archives that Joe
Foote played well the role of a press secretary as he pushed Albert to
take press activities more seriously and to increase his visibility. However,
Albert’s responses to these requests tended to reflect a personal reluctance
to accept media opportunities. Repeatedly, Foote would forward a request
and Albert would decline. To a series of beginning-of-Congress interviews
in 1975 (with major outlets such as the Washington Star, the National
Journal, U.S. News and World Report, CBS Morning News, Agronsky
and Company, and others), Albert handwrote, “I don’t think so.”5 To a
request to be on CBS Morning News to respond to a national speech by
President Ford, Albert declined: “I want to take it easy for the rest of the
week.”6 And, approaching the August recess in 1973, Foote sent Albert
eight requests for interviews from a wide variety of outlets, including print
sources such as Hearst newspapers, the Washington Star, and the New
York Times Magazine as well as television outlets on public television and
ABC’s Issues and Answers. Despite Albert’s increasing interest in media
strategies and the fact that the August recess is a strategically advanta-
geous point in the calendar at which congressional leaders can make news,
Albert’s response to Foote was “No, I am too weary.”7 This is not to say
that Albert declined all invitations, but only that his actual behavior did
not comport with his organizational efforts to engage the media. After a
series of similar exchanges over the years, by June 1975, Foote wrote to
Albert, seemingly resigned to the Speaker’s reluctance to embrace the new
media role: “You have been invited to appear on Face the Nation Sunday.
I will regret this invitation tomorrow morning unless I hear from you.”8
How would political scientists interpret this evidence of contradictory

behaviors and confused or mixed motivations? The tendency among many
would be to search for the “real” or the “primary” motivation. Indeed,
my short answer to questions about Albert’s place in the development
of the public Speakership remains (as it was in the dissertation and the
1998 article) that, like his predecessors and mentors Rayburn and McCor-
mack, Albert was a private Speaker. Whereas this search for clarity and
parsimony is often understandable and laudable, it has the negative conse-
quence of embedding in research a bias against understanding that people
often act without knowing exactly why and, perhaps not infrequently,
do so with ambivalence amidst contradictory impulses. This bias may
pose a challenge to understanding politics in a behaviorally real way.
My extended interpretation—indeed, a chapter’s worth of archival discov-
eries—of this is twofold, having implications not only for Albert’s Speak-
ership but also for questions of institutional development more gener-
ally. On the one hand, Albert sensed the contextual demands for a new
type of legislative leadership but was personally ill-suited and uncomfort-
able supplying it. After a crucial (and tense) meeting with class of 1974
freshmen members and at the prodding of younger, more media-savvy
Democratic leaders (most notably John Brademas) and his press secre-
tary, Albert could read the writing on the wall. On the other hand, despite
his official acknowledgement of the importance of media politics to his
caucus and the House as an institution, Albert was a politician of an earlier
era (and trained well in that era) and he lacked the media experience,
training, and perhaps even instinct that more contemporary politicians—
including his successors in the Speakership—would develop over time.
Perhaps there is a general lesson about institutional change here:

even when environmental conditions change to the degree that there are
demands for institutional innovation, there remains a critical need for indi-
vidual actors to supply that innovation, and some actors will supply that
innovation more willingly, completely, and effectively than others. This
requires an appreciation of the demand and supply sides of institutional
change and innovation as well as an appreciation that institutional change

is not necessarily functional or efficient.9
My point here is not only that archival research allowed me to under-
stand Carl Albert better but also that there is a general lesson: there are
many political questions that are obscured to more traditional political
science researchers that can be illuminated through a closer engagement
with and examination of archival materials. In what follows, I express,
first, the curiosity that archival research has not accompanied the histor-
ical-institutional turn in political science in recent years. Then, shifting
from mere curiosity to overt advocacy, the remainder of the chapter
explores how archival analysis might apply specifically to some of the
more frequent questions that occupy new institutionalist and American
political development (APD) scholars. It concludes with some practical
tips for reconciling ongoing tensions between the ambitions and inter-
ests of institutionalists and those who support APD, on the one hand, and
archival researchers, on the other.
The�Curious�Absence�of�Archival�Research
in�Contemporary�Political�Science
Historically, some disciplinary trends and emphases in political science
have been accompanied by new methods and methodological training
that would diffuse, embed, and normalize new practices—new modes
of research—throughout the profession. One notable expert on archival
analysis (Skemer 1991) traced the twentieth-century history of political
science and the role of archival research in that history.10 In doing so,
Skemer both attributed the mid-century’s decline of archival research to
a rise of behavioralist questions and predicted that in the 1990s, “archival
research by political scientists [would] increase” (367) because historical
and institutional research seemed to be reemergent. Although this was
a compelling and hopeful account that had intuitive merit, it assumed,
among other things, that political science methods follow, functionally,

scholars’ theoretical commitments and research questions.
Inasmuch as it is tempting to characterize disciplinary history as a
progression from the old institutionalism to behavioralism and rational
choice approaches and then again to the “new institutionalism,” these
never have been so much discrete eras as they are convenient, though
incomplete, characterizations of the net balance of the discipline’s rela-
tive affinities for humanism and scientism. Observing the recurrence of
institutionalism across a range of social sciences over time, Dorothy Ross
(1991, 1995), a preeminent historian of social science disciplines, argued
that institutionalism recurs generally as an “opposition movement” and “a
dissenter” to the “scientism of mainstream social science”:
The scientific aspirations of American social scientists have repeat-

edly led them into excessive abstraction, formalism, reductionism.
Institutionalists, drawing on historical values, attempt to restore the
force of historical contingency, the agency of historical actors, and
the contextually based continuities of social relations, ideas, and
values that shaper their fields of study. (1995, 117)
Whereas political science has always dealt with what Ross called a “mixed
allegiance to history and science” (118), the relative mix of these alle-
giances—including the relative strength of institutionalism—has ebbed
and flowed over time. Still, it is fair to say that political science methodolo-
gies, training, and research modes have been less variable and increasingly
have tended to emphasize political scientists’ allegiance to science more
than to history. Perhaps content to remain “opposition movements,” resur-
gent institutionalists did little to develop and diffuse appropriate methods
and approaches that would distinguish their methods from the method-
ologies that are more appropriate to behavioralists and choice theorists.
Indeed, although Skemer correctly identified a significant trend toward
historical institutionalism in the discipline, the predicted return to the
archives failed to materialize, at least to a degree commensurate with
the actual resurgence of historical and institutional questions in the disci-

pline.11 Archival research remains atypical in political science.
This was not always the case. Early political scientists were institution-
alists—the “old institutionalists” of the late nineteenth and early twentieth
centuries—who studied documents, including archives. Importantly, this
was tied to their interest in history, jurisprudence, and institutionalism,
wherein “critical evaluation of historical evidence gathered from books
and manuscripts was still the recommended research method” (Skemer
1991, 359). But, if some of the first political scientists in America were
armed with well-worn historical training and methods in the documentary
analysis of constitutions, official records, and published papers, the behav-
ioral era introduced new questions, and behavioralists invested signifi-
cantly in transforming the methodology of the discipline to match their
questions. Analytically, individual actors, key events, and consequential
(if outlier) facts as potential objects of political study and qualitative study
as a means of understanding and interpreting them had been shunted
aside in favor of mass publics, behavioral patterns, rationalized explana-
tions of those patterns, and quantitative methods and formal modeling
as necessary for comprehending this complex of behaviors and choices.
Thus, political scientists of the mid-twentieth century (whether or not
they thought of themselves as behavioralists per se) were likely to receive
graduate training in the accumulation and manipulation of large data sets
and, for some, in the kinds of survey techniques that are necessary to
analyzing public opinion, constructing national election studies, and the
like. And, when behavioralism advanced, partially gave way, and in some
cases, merely morphed into rational choice analysis in the 1970s, students
were trained in new formal theoretical models and increasingly sophisti-
cated statistical analysis, using computer programs such as SPSS, SAS,
and Stata to facilitate the “new” methods of the “advancing” discipline.
And, when political science departments go looking to hire a “method-
ologist,” this quite typically means they are searching for a quantitative
scholar who is adept in training students in these methods, thus perpetu-
ating behavioral approaches.
Archives, a remnant of the discipline’s past, had been pushed to the

corners by new theoretical commitments, questions, and training. During
the height of the behavioral era, scholars convened at the National
Archives to discuss the state of “invaluable and underutilized” federal
records, and at least one bemoaned the “scanty underpinning of histor-
ical fact upon which many of the published articles” in political science
and public administration journals were based (Hewlett 1975, as cited
in Skemer 1991, 358). Grasping this reality, if hopeful for the future,
Skemer identified several potential explanations for political science’s
dismissal of archives as a potential data source, including a “focus on the
very recent past” (1991, 363), a “fundamental unpopularity of case-study
methodology” (to which one could readily add a preference for quantita-
tive over qualitative forms of evidence), and a downplaying of the role
and importance of institutions in political science in favor of an emphasis
on behavioralism and choice. Still, seizing upon the early trends toward
the revitalization of historical analysis, qualitative methods, and institu-
tionalism that were present and growing in the late 1980s and early 1990s,
Skemer observed that political science might be returning to questions that
would lead back to archival analysis; he counseled archivists to engage
in “aggressive archival outreach efforts” (368) to political scientists and
public policy scholars.
Much has been made of this new institutionalist and historical turn in
political science since the 1980s. In part a challenge to behavioralism
and the simplifying assumptions of rational choice theory, the new insti-
tutionalism, historical approaches, and efforts to “bring the state back
in” asserted the primacy of politics and political institutions and offered
a renewed emphasis on the role of institutions in providing order and
legitimacy, in influencing and shaping the political preferences that were
taken as given and foundational by many rational choice theories, and in
both propelling and resisting change (March and Olsen 1984; Orren and
Skowronek 2004; Skocpol 1995). Although there were several “new insti-
tutionalisms”—including organizational institutionalism, historical insti-
tutionalism, and rational choice institutionalism—the historical variety
served as a foundation for American political development (APD), an

emerging subfield (described sometimes as an “insurgency” [Bensel
2003]) in the study of American politics that focuses its attentions
on diachronic studies of American political institutions and the polity,
culture, and policies.12
If there has been a historical turn in political science in recent years,

there has been little to no accompanying innovation in methodological
approaches and graduate training supportive of that turn. Despite the
very real outreach of professional archivists and the opportunities that
are presented by new institutional questions, the use of archives did not
materialize in political science, at least not at a level commensurate with
the resurgence, on all fronts, of temporal questions, qualitative methods,
and institutionalism. If there is more openness to case analysis (Gerring
2004) and analytical narratives (Bates et al. 1998), the practical means
of data-gathering of detailed historical and organizational evidence has
been lacking. Although I would be the last one to advocate that archivists
stop their “aggressive” efforts to reach out to political scientists, it is clear
that archivists have not been the problem. The failure to connect political
science research agendas with archival data sources likely rests with polit-
ical scientists who seem contented to remain in the dark on some questions
while shining floodlights on other topics, not coincidentally topics where
data—and quantitative data at that—are more easily obtained.
What accounts for this curious lack of historical methods that might
accompany the historical-institutional turn? Why have political scientists
neglected archival sources? As has been chronicled elsewhere in this
volume (in the introduction and chapter 1), in part, it is that they are lazy
and prone to their comforts, archival research is costly, and they lack
training in these methods. In addition to these very real obstacles, there
are three additional reasons, deeply embedded in disciplinary history and
theoretical commitments, that political scientists avoid archival research.
First, political scientists do have different interests, questions, approaches,
methods, and objectives than historians, and even the most “historical”
among them take pains to emphasize those differences. Second, it seems

that as a discipline, political scientists are reasonably content with ongoing
questions and are quite comfortable with the existing (and, as Frisch and
Kelly point out in this volume, easily obtainable) data sources. Third, a
result of all of the foregoing reasons, political scientists remain in the
dark and uncertain about the content of archives and the relevance of that
content to their own research projects and careers. They lack training in
this mode of analysis and as a result, rarely become aware of what answers
may be found in archival sources.
Distinguishing Historical Institutionalism from History

Perhaps one reason that historical institutionalists have failed to jump
at archival opportunities is that they (and APD scholars) have not only
offered themselves up as a “new” way forward in political science,
correcting (as each preceding “new way” has) the omissions and errors
of prior disciplinary trends, but they have also taken considerable pains
to distinguish themselves from historians and the discipline of history.
That is, they explicitly have refused a return to political science’s roots
in history, continuing to stress the importance of theory development,
patterned behavior, and standard operating procedures over the detailed
analysis of institutional workings, key moments, and individual actors.
Concomitant to this, it seems, most have resisted those methods that seem
most appropriate for historians’ questions, even though they might have
some value for their own questions as well.
Satisfaction with Existing Questions

As any American political development scholar might predict, the emer-
gence of historical institutionalism did not wipe the slate clean of the
discipline’s prior commitments, questions, and practices. This is, of
course, as it should be: the behavioral and rational choice eras of the disci-
pline produced many new insights and classic findings and foundations
that unleashed a social scientific approach that stands as a cornerstone for
theory builders and hypothesis testers in the discipline, whether or not they
utilize quantitative methods. And, even if it is valued only as a worthy

opponent, behavioralism and its impulses repeatedly have spawned the
new institutional impulse in the discipline.
Still, despite all of the advantages of maintaining a scientific focus on

behavior and its goal-oriented motivations, this satisfaction is a recipe
for maintaining theoretical and empirical blind spots and for allowing
disciplinary stagnation more generally. The reason is best illustrated by
following a mundane example. When one loses one’s keys, why are they
always in the last place one looks? Besides the seeming irony of everyday
frustration, they are in the last place for logical reasons that are best
considered in light of the two most likely alternatives. First, they are not
in the second-to-last place that one finds them because one continues to
look until they are found. And second, they are not in the “next place”
one might look because one stops looking once one finds them—that is,
locating one’s keys satisfies the quest.
Where do political science researchers look for evidence, and what

satisfies their intellectual quests? First, theoretical commitments have a
self-fulfilling quality. A political scientist who is committed to the idea
that political actors have goals and engage in instrumental behaviors in
pursuit of those goals will either search for evidence of such purposive
behavior until he or she finds it or will simply assume its presence and
build models and report findings accordingly. A paradigmatic belief in
goals sets a researcher in pursuit of a goal-oriented explanation, a pursuit
that is likely not to conclude without finding a set of goals that will
satisfy the researcher and feed the paradigm. This very pursuit precludes
the possibility of discovering and recognizing that behavior is sometimes
aimless, improvised, and rationalized by an actor only after the fact.
Moreover, political scientists come to learn how to match their para-

digmatic and theoretical commitments to “modes of analysis” and data
sets. When might a political scientist (or any scholar) stop looking for
evidence, and how can that affect his or her research outcomes and the
discipline more generally? Participating in a 1995 forum on historical
institutionalism, Morris Fiorina offered a “political scientist’s” critique of

the endeavor as well as an exploration of the positive theory of institutions
(PTI) in rational choice theory. In this account of how a rational choice
institutionalist might approach a research question, Fiorina explained,
first, that he was motivated primarily by an interest in theories of legisla-
tive delegation; second, that he had identified the Interstate Commerce
Act of 1887 as “the first major delegation of congressional regulatory
authority” (111); and third, that he began looking for evidence of policy
makers’ considerations of the logics of delegation:
For several years I worked almost full-time on this project. I read

articles published in the nineteenth century and books that had not
been checked out since the 1930s (no big deal for most of you, but
I’m a political scientist!). I copied and read more than 1200 pages of
the Congressional Record. I hired RAs and mapped economic char-
acteristics onto the congressional districts of the 1870s and 1880s.
I did some preliminary statistical analysis. And then I quit. (112)
He said he quit the analysis when “it became apparent…that delegation

logics didn’t have much to do with passage of the ICA” (112) and because
he believed “the story was interesting mainly to the extent that it shed
light on the particular theoretical idea that motivated the study” (112);
apparently reality was found wanting.
This fine scholar’s admission that the lack of evidence for his theory
led him to drop the project altogether is, of course, but one potential
response to such a pattern of research events. Other scholars, motivated
more by an understanding of the Commerce Act itself, might drop the
theory or search for a more appropriate explanation from another theoret-
ical perspective. Still others might look for a refinement of the theory, an
explanation as to its seeming irrelevance to the case at hand, or even seek
to induce a new theoretical model. If scholars, driven primarily by theo-
retical commitments, look elsewhere when data sources fail to provide
confirmation, then they never question the substructural aspects of their
theoretical commitments. Thus, a commitment to the rational choice para-
digm sets scholars looking for rational, interested motivations for polit-
ical behaviors (often leading them to continue to look until such motiva-
tions are found or simply to assume them when evidence of motivation is
unknown or unknowable). Once they are satisfied that a “rational” expla-
nation has been identified, scholars are apt to look no further for a poten-
tially contradictory explanation or additional factors that would compli-
cate explanation and analysis. Holding a primary theoretical commitment,
then, conveniently reinforces the overall theoretical project. It is, after all,
easy to develop, promote, and retain timeless covering laws, as some have
an ambition to do, if one is willing to consider only the cases from history
where a confirmation of one’s prior-held theoretical commitments is easily
obtained.
Continued progress in the broader discipline, however, requires that no

one set of commitments predominates and that political scientists resist
being so quickly satisfied by their answers. Instead, they should ask,
What more can be known? And, how can we begin to know it? Although
political science benefits from having a variety of different scholarly
tastes, skills, and dispositions, any tendency or bias in the discipline
toward one set of tastes, skills, and dispositions can have broader conse-
quences—positive and negative—for understandings of political institu-
tions, behavior, and development.
Uncertainties Surrounding Archival Research

If the self-critical search for “What more can be known?” is good for
any individual scholarly endeavor, it is especially critical for the disci-
pline as a whole. What questions have not been considered? To what
answers have scholars closed their eyes as a result of their theoretical
commitments, research methods, and satisfactions? As a discipline, polit-
ical scientists could be more open to theoretical and methodological
pluralism and to retraining as questions and norms change over time. In its
efforts to train graduate students and develop methodological norms, the
behavioral “revolution” in political science was more complete and self-
sustaining than new institutionalism seems to have been. Graduate courses
in scope and methods and the textbooks that are used in those courses
have embraced behavioralism, taught its skills, and in doing so, prolifer-
ated and sustained its practice.13 And, despite some resurgence in qualita-
tive methods training, graduate programs’ research methods courses and
political science methods textbooks continue to emphasize the kinds of
sophisticated, quantitative methods and formal modeling that were devel-
oped for political science’s behavioral era. Lacking training in archival
methodologies, political scientists have been loath to venture outside their
methodological comfort zones.
Political scientists’ reluctance to engage in archival research seems to be
a consequence not only of this lack of training but also of a general uncer-
tainty as to the content of such collections and the relevance of that content
to their questions. Will they find their answers in the archives? Indeed,
as Frisch and Kelly correctly observe in chapter 1, a negative answer to
this question involves not only a loss of precious time but also poten-
tially the significant financial investment that is attendant to traveling to a
collection or, more likely, numerous collections spread across the country
to obtain evidence and data. Uncertainty surrounding this question and a
general lack of models in the literature of how political scientists have used
archives likely has dissuaded some scholars from making the attempt. Just
as likely, the disciplinary norms being what they are, many never even
consider archival research as a viable means of gathering evidence.
To be sure, archival methods are not of equal utility for all political
science questions. At one level, it depends on the political scientist and his
or her questions. Political scientists focused on theoretical questions who
are looking for institutions and data sets to test and explore those commit-
ments may, like the Fiorina example earlier, strike out and end up with a lot
of costly paper and considerable wasted time. But for political scientists
whose primary commitments are not theoretical but substantive—that is,
they see theory not as the end but as the means of understanding political
institutions and behaviors—archival research is much less of a gamble. If
the primary interest is in how the Congress or the presidency develops,
how congressional incumbents campaign, how interest groups strategize

to influence public policy making, how the public opinion industry devel-
oped and worked to legitimate itself, or any other of a myriad of substan-
tive questions, documents waiting in archives offer important opportuni-
ties for insight.
A�New�Mode�of�Analysis�for�Political�Science’s
Historical-Institutional�Turn
Archives can offer documentary evidence of the kinds of complex deci-
sion-making and strategic behaviors of political actors that conveys the
behavioral reality of political actors working in real time and in contexts of
uncertainty. Moreover, leveraging historical evidence and data situated in
institutions, archival analysis offers the possibility of tracing institutional
change intimately, as a historian might, but with a political scientist’s eye
toward theory development and testing. This is all to say that evidence
in documentary sources meshes well with the questions that occupy new
institutionalists and APD scholars.
Archives and the New Institutionalism

Institutions produce paper and lots of it. The process of “institutionaliza-
tion” itself often involves the creation of organizational scripts, suborgani-
zational structures, standard operating procedures, and the offices, forms,
and procedures that allow such a complex institution to function with
predictability, efficiency, and accountability. Moreover, officialdom and
record keeping are on the side of those who wish to understand public
institutions better because the kinds of official records that are retained for
the purpose of public accountability are also maintained for posterity, and
many eventually are opened for academic research. Until the most recent
eras of computing and computer-facilitated communications, paper has
been essential to the internal workings and the external monitoring of insti-
tutions. Paper trails are forged by even quasi-official organizations and
can be an early sign of organizational innovation and change. Indeed, once
when I was trying to convince a colleague of the importance of two party-

message organizations—the Democratic Message Board and the Repub-
lican Theme Team—it was only when I mentioned that they had letter-
head that this colleague seemed convinced that these organizations were
regular, ongoing, and important suborganizations in the House legislative
parties and worthy of study. And, when I turned my attention to analyzing
the political composition and activities of these organizations, it was the
listing of all of the Republican Theme Team members on letterhead that
facilitated my (2008) analysis of what types of members participated in
these message organizations, so central to House party communications.14
Here and elsewhere, letterhead, meeting minutes, expense accounts and
receipts, working papers, strategy documents, and memoranda are institu-
tional paraphernalia that exist as evidence of the actual workings of orga-
nizations and their actors.
Intensive study of these documents meshes well with the questions

that new institutionalist scholars pose and that have long separated them
from noninstitutionalists in multiple disciplines. Scott—making explicit
reference to Jacoby (1990)—cited four key points of departure between
early institutionalists in economics and those of the neoclassical school
as fundamental debates repeated across the social science disciplines,
including political science. These ongoing debates center around ques-
tions of “indeterminacy versus determinacy,” “endogenous versus exoge-
nous determination of preferences,” “behavioral realism versus simpli-
fying assumptions,” and “diachronic versus synchronic analysis” (1995,
4).
First, indeterminacy versus determinacy: The search for determinant

outcomes, functional markets, and efficient progress toward equilibrium
points occupies neoclassical economists and the choice-based scholars
they influence in other disciplines. Reading history back, it is easy to
develop a post-hoc, rationalized story of how institutions and practices
developed because they were needed, because they served some institu-
tional actor’s strategic aims, or simply because their time had come. But
viewing developments from the perspective of individual actors operating

in real time—as archival research allows—points up the uncertainty that
is inherent in most moments of institutional flux and upheaval as well as
the failed attempts, false starts, and trial-and-error learning that are atten-
dant to any new institutional form or behavioral practice. Archives help
scholars appreciate the difficulties associated with institutional change,
the uncertainty and ambivalence of change agents, and the overall messi-
ness of historical progress in real time.
Second, endogenous versus exogenous determination of preferences:
How individual political actors’ political preferences are shaped is a
question that occupies scholarly debate, although this debate has gener-
ally been determined more as a matter of paradigmatic preference than
empirical investigation. Some scholars express faith that basic human
motivations—goals—are more or less fixed, and they build their models
of political behavior and institutional analysis accordingly. Others see
goals as malleable, affected by ideas, imparted by institutionalization,
and sustained by norms or constrained by institutional rules, procedures,
and processes. Some, following E. E. Schattschneider’s famous claim
that “people have ideas about interests” ([1942] 1970, 37), believe that
identity, ideology, and even personality affect how individual political
actors perceive, perhaps if only to rank-order, their multiple and poten-
tially contradictory interests. Others, including some rational choice insti-
tutionalists, believe a political actor’s preferences may also be affected
by his or her political context, including where he or she may be situated
in an institutional setting. Generally, one needs to appreciate that given
the multiplicity of preferences that an actor might have, he or she may
find a particular moment, decision, or institutional arrangement more or
less hospitable to the expression of one of his or her preferences over
others. Here, too, one finds that the institutional arrangement has played
an important role in shaping the preferences that are at work in a particular
political circumstance. And, sometimes, individuals may want to advance
goals but find such an advance likely to be futile in a particular institutional
context. What role, if any, do individual actors and events have amidst
the wash of other factors that are already embedded in and operating on
a political system as well as the inertia that is inherent in many political
institutions? These are complex problems that continue to occupy scholars
across the social sciences, and the debates over these problems have been
allowed to remain a matter of taste and theoretical preference rather than
being subjected to empirical tests. To the extent that archival documents—
correspondence, personal notes and diaries, and/or strategy memoranda—
allow scholars to see consequential political moments through the eyes
of individual political actors or as groups of such actors seek to coordi-
nate, reconcile, and advance their common goals, evidence uncovered in
archival collections can help them better understand whether and perhaps
how the preferences of some political actors are shaped and how the
contexts in which those actors act and choose might both foster and
constrain the expression of those preferences.
Third, behavioral realism versus simplifying assumptions: Perhaps the

true advantage of archival analysis is that it offers a view as to how
politics actually works and, when the data are rich, how political actors
respond to (and plan in light of) complex, changing, and uncertain envi-
ronmental cues. Again, as with the determination of preferences, when
scholars are data poor, they fill in the blanks, either letting theory guide
(sometimes to the point of tautology) or allowing author imagination and
interpretation to reign. As a social science, political science would be
wise to check its theoretical ambitions and assumptions against actual
evidence from the empirical world. Do politicians behave in practice as
scholars expect them to do in their theories? Politicians are often apt
to hide behaviors—some behaviors quite a bit more than others—and
archives hold the possibility of pulling back the curtain and allowing
observers a closer look at how politics actually happens. For years, polit-
ical communications scholars have argued that political actors engage
in strategic message framing and that the conveyance of those message
frames has effects on mass opinion and behavior; still, there remained very
little evidence that politicians “framed” messages in the self-aware and
strategic way that political scientists’ studies have attributed to them. My
(2010) archival research in congressional party communicators’ papers

confirmed not only that Democratic and Republican legislative leaders
alike did engage in multiple framing efforts but also that they under-
stood framing strategies and self-consciously used the term framing when
considering their actions. What if the alternative had been true? It is just
as easy to imagine that political framing is epiphenomenal of years of
embedded political-speak rather than the learned and coached behavior
it actually appears to be. In any event, nailing down such expectations
solidifies scholars’ research projects and findings and provides a window
on how politicians actually behave, standing all the while as a potential
corrective to mistaken impressions and faulty assumptions.
Finally, diachronic versus synchronic analysis: Institutionalists are

more likely to engage in developmental and comparative-historical (that
is, diachronic) analysis rather than to assume or even search for universal
and timeless theories of politics. Diachronic analysis includes the use of
the “laboratory of history” (Cooper and Brady 1981b) for the purposes of
testing social science theories, often in a comparative way, across eras.
These within-country-but-across-eras comparisons provide snapshots of
how the same political system or institution operates under different
conditions and at different points in history (Burnham 1982; Cooper and
Brady 1981a). It is in these kinds of studies that the causes and conse-
quences of political change are likely to be in the sharpest relief. And,
more generally, scholars deem as diachronic those studies that investigate
“how [a system] acquired its features and the conditions that cause these
features to vary over time” (Jacoby 1990, as cited in Scott 1995, 4). Both
kinds of diachronic analysis require a flexible data set in terms of the time
scope, better and more detailed evidence within eras, and more generally,
an elaborate appreciation of an institution’s context on multiple dimen-
sions, including as those multiple dimensions might also change over time.
To the extent that institutionalists are interested in how things change,

how preferences form, how politicians actually behave, and historical
development and comparisons, a deeper appreciation of historical data
methods is warranted. Of course, political scientists’ scientific aspirations

counsel against relying too much on interpretive methods and interpre-
tivist studies, but all scholars—formal and quantitative included—embed
assumptions and overinterpret the data that they identify to the exclu-
sion of possible alternatives. Political scientists are all interpretivists; if
it is true that some are more interpretivist than others, it is also true that
some hide it much better than others. There is no denying that archival
researchers necessarily engage in interpretation to advance claims about
behavior and institutional performance (and may do so more than many
of their more “scientific” colleagues do), but those interpretations can be
supported by documentary evidence from the politician’s point of view. To
the extent that scholars offer their readers evidence derived from archival
materials, the readers can judge the veracity of the scholars’ claims and
the wisdom of their assumptions in light of the evidence they marshal.
Such a “look and judge for yourself” presentation of data would go a long
way to making interpretations of individual political actions and broader
expectations of institutional change more grounded in empiricism and less
reliant on mere theory.
Archives and American Political Development

Although APD scholars have an interest in all of the foregoing questions
regarding institutionalism, it is to questions of “diachronic analysis”—the
processes, causes, and consequences of institutional change (and change
in the broader polity more generally)—that they direct the most attention.
American political development emerged in the 1980s, consistent with—
perhaps an offshoot of—the new institutionalism’s interest in the historical
development of institutions.15 It is the most self-consciously organized
historical movement in political science since the beginning of the disci-
pline.16 But, for all of this self-consciousness, this “historical turn” has not
been accompanied by a greater appreciation, practice, and dissemination
of historical methods.
APD took the impulse of the new institutionalism in a more deeply

historical direction, focusing on the context of American politics. In full
disclosure, I teach APD and I consider its lessons and research accom-
plishments to be among the highest order of complex and innovative
thinking and theorizing in political science. Moreover, it is full of gems
of insight, anecdotes, and interesting fact patterns that enliven the mind
and the classroom. In short, I am a fan of APD scholarship (though, as
will become obvious, not an uncritical fan). I would argue that although
it was suspected to be a mere trend in the late 1980s and early 1990s,
APD has been a success—and a sustained one at that—with the potential
to contribute to the ongoing historical transformation of political science
with no less impact than the old institutionalism or behavioralism. And, as
Dorothy Ross might add, it has been a powerful corrective to the reductive
influences of some of political science’s other theoretical bents.
Still, APD has had its blind spots (both substantive and methodolog-
ical), and there are lingering question marks about the future of the
subfield. First, as a Congress scholar, I willingly join in the critical obser-
vation that APD has underappreciated the role of the legislative branch,
favoring instead a focus on administrative politics, courts, party poli-
tics, citizenship, and state development more generally (see Katznelson
and Lapinski 2006). Of course, this is a criticism largely about focus
rather than method. Still, it is quite ironic in that to the extent that APD
scholars are focused not solely on political development but on a uniquely
American political development, the singular importance of Congress in
America (compared to legislative power elsewhere in the world) invites
a closer consideration of the role of Congress in promoting American
exceptionalism (see Roof 2011) and in shaping the American polity more
generally.
Second, and more fundamentally, APD’s grand ambitions for broad

theorizing and doing so in a wide historical scope sometimes strain its
ability to produce detailed historical work and empirically grounded,
behaviorally real accounts of institutional development. Not only do APD
scholars engage large swaths of time in their studies, but they are apt
to stress the cross-institutional nature of their work as well; some even
make such a cross-institutional focus a differentiating characteristic of

APD work. To be sure, when added to APD’s commitment to studying
historical change, this challenges the scholar’s ability to provide behav-
iorally real, nuanced, and detailed accounts. At the same time, however,
according to Orren and Skowronek,
the preference in the APD literature is for “thick” over “stylized”

description of [historical] sites, descriptions that show an array of
different pieces and how they are associated and also for process
tracing—examination of the reverberations of a change and of
carryovers that are largely unaffected by it. (2004, 21)17
But just how “thick” are the descriptions that APD offers, and how
“thick” can they be if APD remains committed to broad historical scope,
grand theorizing, and cross-institutional development? Some scholars
seem to regard APD historical accounts (see Gerring 2003) as not “thick”
enough, calling both for a greater appreciation of descriptive work and for
greater specification of concepts and causation in APD studies (Orren and
Skowronek 2004, 96–97). Even if they are less “stylized” than rational
choice accounts, many APD accounts necessarily forsake detailed histor-
ical accounts for a focus on broad historical theorizing. Compared to histo-
rians, APD examinations often involve glossed interpretations of actual
events and historical processes.18
Third, whatever APD’s focus and ambitions have been, it largely has
conducted its research (again, let me stress that there are important excep-
tions) without resort to archival research, leaving most of the detailed
work to historians. If this is generally true of how political scientists
use history,19 interestingly, APD scholars, too, rely on historians for
more detailed analyses of events and actors and, problematically, treat
those historical accounts as detailed fodder—even “fact”—for their theo-
rizing.20 Although history often provides sufficient resources, this reliance
nevertheless subjects the quality of scholars’ analyses to the trends and
dispositions of historiography (see Lustick 1996). And, to the extent that
they appreciate the long-held differences between the discipline of history
and that of science, political science cannot rely solely on the disciplinary
outsourcing of its historical data-gathering.
History, however, is not simply the event analysis, case studies, and fact
patterns that political scientists seem to treat it as. As Lustick correctly
observed,
The work of historians is not understood by historians to be,

and cannot legitimately be treated by others as, an unproblematic
background narrative from which theoretically neutral data can be
elicited for the framing of problems and the testing of theories.
(1996, 605)
Instead, “background narratives are constructed, not discovered, and…

they carry theoretical and perspectival commitments which raise signif-
icant problems for, and can invalidate, the use of history as a labora-
tory for social science” (Lustick 1996, 613). Understanding this, political
scientists must appreciate the costs of off-loading their historical work
to others. When possible, why should they not do some of the historical
work themselves? To be sure, constructing historical accounts will be no
less interpretive when it is political scientists who are constructing the
historical narrative, but both the author and the reader will necessarily
be confronted with that construction in process. An author using archival
materials to uncover an account of events, behaviors, or institutional
developments should be expected to present the evidence—describing and
perhaps quoting from the sources—as well as to document and contextu-
alize the evidence in a way that equips the reader to judge the value of the
evidence and its ability to shed light on a problem or to resolve a ques-
tion in a dispositive way. Political scientists’ current tendencies to cite a
historian’s account (and even a contemporaneous news account) as fact,
though it is understandable and often harmless to arguments and truth,
nevertheless tends to render authors and readers alike less aware of the
embedded and hidden interpretations that, directly or indirectly, went into
the selection of a particular historical source and/or the construction of
a historical account that is taken—that is, mistaken—for a true and unaf-

fected chronicling of events.
For political scientists to resist this outsourcing to history would require
that they become more willing to admit of more historical, descriptive, and
archival work. The discipline would have to place a higher premium on
the development of “data” and information that inform scholars’ under-
standings—however derived—of politics. Whereas Gerring observed that
political science’s interest in history relies quite a bit on the traditional
writing of historians, he also argued that this might be changing and that
political science would be wise to prepare for the likely “eventuality” that
political-history writing will “expire”:
A return to grand history within political science, therefore,

presumes the survival of a traditional style of political history-
writing that covers the ground more closely. Traditionally, this
case-study work has been provided by historians. Should this
vein of history-writing expire, however—an eventuality that seems
increasingly likely—it will have to be resuscitated by political
scientists. This, in turn, will necessitate the revival of closely-
focused, nose-to-the-grindstone efforts. There is no sense in turning
to grand synthesis before some sort of empirical groundwork has
been laid. Thus, the direction of political history within political
science depends very much on the state of political history within
history. (2003, 102)
Ultimately, even if balancing the twin impulses to cover more institutional

and historical ground and to do so in a more detailed way proves diffi-
cult for any particular study, APD (and political science generally) would
nevertheless be wise to continue both its high theoretical ambitions and its
inclinations toward better understanding the behaviorally real in American
government and politics. By balancing its interest in its paradigm-pushing
grand theory with an equally necessary paradigm-filling detailed empir-
ical work, APD could push ahead on both fronts.
This would require, however, that APD resolve its identity crisis, even if
by embracing it. Caught between the ambitions of some scholars to trans-
form political science as a discipline and the desire of others to maintain
the subfield’s “insurgent” quality (Bensel 2003), APD leaders are uncer-
tain of how to build upon their successes. In response to Gerring’s (2003)
recommendations that APD normalize its questions and approaches in the
discipline (by, it seems, integrating them with more generally accepted
political science methodologies), Skowronek wrote:
APD has been a great success in several respects. The question, it

seems to me, is whether in the wake of this success it can retain
a distinctive voice…What previously passed for an agenda is now
common wisdom…All this leaves the field at what practitioners
themselves might recognize as a critical juncture. Absent more
careful consideration of what the APD project is—of what distinc-
tive and collective contribution its practitioners have to make to the
understanding of politics today—it stands simply to dissolve away
or become absorbed into everything else. (2003, 110)
No lesser leaders of the APD movement than Bensel and Skowronek, then,
seem to see the stakes of this normalization as more of a risk than an oppor-
tunity. Where others might see the improvement of political science’s
understanding of institutions and political change as a paradigmatic shift
that could improve the understanding of politics generally, these scholars
seem to be just as interested in protecting their insurgency as they are in
normalizing and proliferating their perspectives. Without a reconsidera-
tion of this, it seems likely that the end result of this identity crisis will
be that APD will remain content to correct, as institutionalists have in the
past, the excesses of the discipline’s scientist aims (Ross 1995).
If, however, APD pursues the ambition of some to transform the overall
disciplinary orientations, then this effort will be incomplete until polit-
ical scientists train graduate students and colleagues in new methods and
modes of political analysis. Could archival methods be the future of APD?
Why not? Should they follow Gerring’s advice that they engage in more
“nose-to-the-grindstone” historical research, political scientists’ appetite

for archival sources might, at long last, increase. Archival resources,
heretofore largely the historian’s domain, offer a number of potential bene-
fits for APD scholars. For example, as Congress (hopefully) becomes
more integrated into the APD project, there is a great abundance of
archival resources that can help chart the development of Congress as an
institution, demonstrate its complex interactions with other branches and
interplay with the broader polity, and show its primary role in the devel-
opment and oversight of the public policies that shape America (Arnold
1982). Beyond the study of Congress, there are several potential applica-
tions of archival sources and methods to questions typical of APD. Indeed,
this criticism about the lack of attention to Congress notwithstanding,
even APD’s initial focus on administrative politics and the development
of the state as well as the origins and consequences of public policy
seems to represent an opportunity missed, because other scholars have
found archival resources particularly useful in policy studies, particu-
larly from the perspective of administrative agencies (Skemer 1991, 358).
By the same token, Amy Fried’s exploration (in this volume) of how
archival research has accompanied the “historical turn in public opinion
research” is but one example of how archives can be put to this use. Tracy
Roof’s deep engagement with labor union archival sources (also in this
volume) sheds important light not only on union politics and labor strate-
gies but also on congressional politics and the centrality thereof to Amer-
ican policy making and welfare state development.
So long as APD regards as central to the endeavor their insurgent

identity freed from disciplinary norms or the kinds of cross-institutional
analysis that bridge state and polity or multiple institutions, it may well be
that rigorous primary research is untenable (Lustick 1996, 606; Skocpol
1984, 382). What would make APD’s grand ambitions sustainable would
be a duality of effort both to produce theories and to test those theo-
ries across institutions and eras. Importantly, doing both well is likely to
involve APD scholars engaging in the kinds of detailed work of archival
analysis that are in evidence throughout this volume. What better future
for American political history could there be than to adopt a historian’s

tools but nevertheless retain a political scientist’s eye and appetite for
theory?
The Kinds of Evidence Archives Have to Offer

Scholars who are interested in behaviorally real, behind-the-scenes infor-
mation concerning actors and institutions as well as in how those actors
perceive institutional context even as it changes over time will find in
archives a mountain of evidence to feed their research agendas. In my
own research in the documents of congressional leaders, I have iden-
tified several types of documents that I believe might fit into broader
categories of documentable actions and behaviors across institutions and
prove useful to many different political scientists, especially historical
institutionalists and APD scholars. That is, if the examples below are
Congress-specific, the broader categories are likely to be in evidence
(though to different degrees) in the archival holdings of officials and orga-
nizations in the executive branch, the judiciary, interest groups, political
party organizations, public opinion and media organizations, nongovern-
mental organizations, and individual actors. These documents, after all,
are the work product of politics in action.
First, archival materials contain memoranda, letters, and correspon-

dence that connect an institution’s actors to one another and to those other
relevant actors outside the institution with whom they interact. In terms
of the documents uncovered in congressional leadership archives, exam-
ples include the wealth of “Dear Colleague” correspondence that members
of Congress use to make policy statements aimed at coalition building
and persuasion and to discuss internal legislative procedure and politics
as well as less formal, ad hoc correspondence between two members, a
member and a legislative leader, two legislative leaders, or even between
legislators and the president or other actors in the White House. These
latter documents provide insider information on behind-the-scenes nego-
tiations and relationships that are important, if sub rosa, aspects of internal
congressional politics. Among the richest types of institutional corre-
spondence are letters and memoranda between principals and staff or

among staff members within the same congressional office. One notable
example of this can be found in the papers of Republican House Leader
Robert Michel (R-IL); documented among this large collection are interof-
fice memoranda between Michel’s leadership office and his personal and
district offices in Washington and Illinois, respectively. The sometimes
contentious disagreements among his own staffers over how to improve
Michel’s image and balance his multiple responsibilities demonstrate well
the puzzle that confronts House leaders as they balance their personal,
party, and institutional responsibilities (on this, see Green 2010). More
generally, these kinds of deeply behind-the-scenes correspondences offer
the kinds of frank exchanges that allow a researcher to follow along as
a political enterprise (Salisbury and Shepsle 1981) seeks to identify its
weaknesses and solve organizational or political puzzles.
Second, archives contain the kinds of planning and strategy documents

that shed light on how real-life political actors perceive often complex
and uncertain contexts and make decisions and choices of action in light
of those perceptions. One advantage of this elite actor-centered investiga-
tion is that scholars might be able to see the political or institutional land-
scape—the complex of multiple and potentially contradictory contextual
factors—as their research subject and his or her staff see it. Instead of
relying on theoretical, hypothesized relationships between one or a few
contextual factors and an actor’s choice, scholars benefit from seeing the
political terrain through a politician’s eyes—at least as those perceptions
might be articulated in documents—as the politician reveals his or her
professional political sense of what factors might affect such complex
decisions. Data, reports, arguments, and opinions supplied to decision
makers by constituents, policy experts and think tanks, interest groups,
and other policy makers are frequently in evidence in archival holdings
and offer the researcher a more complete sense of the various kinds of
information that are available to decision makers and the dilemmas that are
produced by some complex choices. Of course, a wise researcher will be
aware both that not all that a politician considers in making such decisions
is in document form and that not all documents that are possessed and
retained by an office have been considered equally or seen or considered
at all by a member of Congress, a president, or any political actor. Still, the
sometimes vast range of available information provides a scholar with a
better sense of institutional context and might even introduce that scholar
to a whole other political or policy dimension than he or she had previously
considered. As with participant observation (see Fenno 1986), developing
closer contact with research subjects by examining their papers can reveal
more about how politics actually happens.
Third, documents in manuscript collections include organizational arti-
facts and work products. In addition to the official records of a congres-
sional committee, a caucus or conference meeting, or the legislative
floor itself, there are many less-formal organizations that operate within
Congress, playing important and consequential roles. The quasi-official
documents of these organizations provide the institutional researcher with
detailed information that would otherwise be unavailable. From Carl
Albert’s Information Task Force to Newt Gingrich’s Conservative Oppor-
tunity Society, serious congressional innovation and development begin
in informal quasi-organizational settings that are not “official enough” to
require regular reporting (by law or congressional rule) but are both conse-
quential enough to merit scholarly attention and formal enough to take on
the organizational processes of attendance, taking minutes, and formal-
izing divisions of labor in written form. To be sure, these congressional
leadership examples have their analogues in informal working groups and
task forces operating in other political institutions, the kinds of venues
where policy makers and those who seek to influence them brainstorm,
guess, debate, measure, reconsider likely political and policy outcomes,
and speak frankly (even as they are developing but a nascent sense) about
their political objectives and circumstances. Such evidence allows insti-
tutionalists to flesh out a historical record that is likely too hidden and
detailed to occupy some political scientists but too organizational, stan-
dard procedure, and “inside baseball” to attract a historian’s attention.
Still, for those scholars who want to know how Congress, the White
House, or judicial chambers actually work day to day, these documents

can be a gold mine.
Fourth, as Frisch and Kelly suggest in chapter 1, archives contain
the kinds of behind-the-scenes data and analysis that even quantitative
political scientists could mine and explore. Like many other institutions,
Congress is awash in information about policy and politics. Whether it
is produced internally (that is, by the institution itself) or supplied to the
institution by friendly and interested actors with whom they interact, this
information provides a window on the relative balance (and complexity)
of choices and tensions that are inherent in any political decision. When it
is viewed in light of the subsequent authoritative decisions that are made, it
might provide a clearer picture in any political or policy battle of who won,
how much was won, and why. For example, it may be true that a certain
policy enjoys a high degree of public approval in published polls, but
uncovering the poll results that an individual decision maker was using can
shed new light on how he or she perceived the public opinion context as
well as the amount of detailed cross-tabulations and analytic metrics that a
politician or a party deemed more important than just published approval/
disapproval figures for making a particular decision. The data that polit-
ical actors peruse, collect, and produce is rarely subjected to the kinds
of rigorous hypothesis-driven analysis that social scientists engage in,
whereas these same kinds of “finds” that would aid a historically oriented
scholar’s analysis would otherwise be lost to the historical record. Gath-
ering this information is a significant opportunity for political science’s
scientific and historical impulses. By the same token, even more publicly
available documents such as the transcripts of leadership press appear-
ances or their press conferences, which may or may not be collected or
readily available to scholars, are often located in one place in a leadership
archive, thus making an archival trip well worth it whether the researcher
hopes to quantify or contextualize with rich descriptions.
In these and other respects, the evidence, information, and data

contained in many archival collections exist to address critical historical
and institutional questions confronting political scientists. No advocate

can honestly promise that a scholar will find his or her answers in a
particular archival site. But scholars who are looking to drill down a bit
deeper than examinations of the official records of the American national
government, those who are in search not only of officialdom but also of
the behind-the-scenes, behavioral reality of how organizations actually
work, will find important insights and evidence through sustained archival
research across several repositories.
Conclusions
Political scientists’ avoidance of archival analysis has been a serious and
costly mistake, particularly as the discipline has returned to institutional
and developmental questions. New institutionalists and scholars of Amer-
ican political development would deepen both their understandings and
their analyses through a more sustained and intentional engagement with
archival resources. And, even if more of these scholars do so in their own
work, political science’s historical turn will not be complete until polit-
ical scientists train their students and colleagues in appropriate historical
methods and those methods appear alongside quantitative methods and
formal modeling on graduate programs’ scope and methods syllabi. Is it
not a little ironic that the behavioralists were better at building the insti-
tutional support and sustenance for their paradigmatic projects than the
institutionalists were?
I do not mean to suggest that only archival methods can unlock the ques-
tions that occupy historical institutionalists and APD scholars. To be sure,
scholars’ methods and data must be appropriate to their questions. No one
is as aware of the limitations of archival methods as is a practitioner of
those methods. Scholars engaging in archival research must endure the
sometimes slow plodding of the endeavor and the not infrequent frustra-
tions due to the incompleteness and unevenness of the archival record.
Still, despite these problems, archival methods remain a potent source of
information and inspiration for scholars who take organizational struc-

tures and their impacts, behavioral reality, and institutional change seri-
ously. My hope is not for a new methodological hegemony centered on
archives but instead for a proliferation of the kinds of theoretical and
methodological eclecticism that are necessary to understand more about
something as complex, multifaceted, and human as politics and govern-
ment. My advice to colleagues, then, is to read on in this volume with
both a curiosity for what else political scientists might be able to learn and
know and with an active interest in exploring how the evidentiary discov-
eries and analytical advances of these fine scholars might be analogized
to and replicated in their own research questions.
Endnotes
1. On the concept of modes of analysis, see Fenno 1986; I have addressed

some of these issues previously in Harris 2009.
2. For a contrast of these different modes of analysis, see Harris 1998 and
Harris 2005.
3. This is an appropriate place to acknowledge the good work and assis-
tant of archivists at the following collections to which I have traveled,
which inform this chapter, and some of which are cited in the notes in this
chapter as appropriate. Thanks to the archivists at the following collec-
tions: the John McCormack Papers, Special Collections, Boston Univer-
sity (JWM); the Carl Albert Papers at the Carl Albert Research Center,
at the University of Oklahoma (CAC); the Thomas P. O’Neill Papers,
Special Collections, John J. Burns Library, Boston College, Chestnut Hill,
Massachusetts (TPO); the Jim Wright Collection, Mary Couts Burnett
Library, Texas Christian University, Fort Worth, Texas (JCW); the Papers
of Thomas S. Foley, Special Collections, Washington State University,
Pullman, Washington (TSF); the Robert H. Michel Collection, Dirksen
Congressional Research Center, Pekin, Illinois (RHM); the James H.
Quillen Papers, Archives of Appalachia, East Tennessee State University
(JHQ); Records of the House Democratic Caucus, Manuscript Division,
Library of Congress, Washington, DC (HDC); the Richard A. Gephardt
Papers, Missouri Historical Society, St. Louis, Missouri (RAG); the
Richard A. Armey Papers, Carl Albert Center, University of Oklahoma
(RAM); John Brademas Papers, Office of University Archives, Elmer
Holmes Bobst Library, New York University, (JB); the papers of Eddie
Boland at the John J. Burns Library, Boston College, Chestnut Hill, Mass-
achusetts (EB); the papers of James G. O’Hara at the Bentley Histor-
ical Library, University of Michigan, Ann Arbor, Michigan (JGO); the
papers of Sam Gibbons, Special Collections, University of South Florida
(SG); the papers of Frank J. Thompson at the Seeley G. Mudd Manuscript
Library, Princeton University (FJT); the Morris K. Udall Papers, Special
Collections, University of Arizona (MKU); and the Papers of Represen-
tative Newt Gingrich, Special Collections, University of West Georgia
(NLG). I am pleased to acknowledge the support of the Everett McKinley
Dirksen Congressional Research Center and the Caterpillar Foundation,
the Carl Albert Center, and Loyola University Maryland, which supported
my travel to archives and interviews with congressional staff. Thanks

also to Speaker Newt Gingrich and the staff at http://www.newt.org, who
allowed me early access to the Gingrich papers.
4. Joe S. Foote memorandum to Carl Albert, 1/9/73, Speeches Box 18,
Folder 6, CAC.
Folder 59, CAC.
6. Joe S. Foote memorandum to Carl Albert 5/27/75, Speeches Box 20,
Folder 60, CAC.
7. Albert handwrote this response on Foote’s 8/3/73 memorandum, Speeches
Box 18, Folder 7, CAC.
Folder 60, CAC.
9. On the first point, see Gamm and Shepsle 1989 and Swift 2002; on the
second point, see March and Olsen 1984.
10. I strongly encourage the reader to read Skemer’s (1991) work on this topic
for its detailed analysis of the evolution of political science and the place
of archival research in that evolution. What follows in this section is a
critical examination and extension of that excellent source.
11. This is a good place to acknowledge that scholars, including many of the
authors in this volume, have returned to the archives. Still, the number of
those who have done so is small and the opportunity costs to the discipline
are, I believe, significant.
12. Even its top practitioners differ on exactly what American political devel-
opment is and should be; see Gerring 2003, Bensel 2003, Skowronek
2003, and Smith 2003. Still, an excellent statement of some of the aims
of the subfield (as well as an acknowledgement of the debates swirling
around those aims) can be found in Orren and Skowronek 2004.
13. Even at the more qualitative graduate program where I did my PhD
studies, the research methods courses focused almost exclusively on quan-
titative methodologies.
14. Although the membership of the Theme Team in the 106th Congress was
publicly available, complete rosters were unavailable for Democrats or
for Republicans in other Congresses. Several Democratic Message Board
and Theme Team rosters were uncovered in archives; the membership
list of the Democratic Message Board in the 101st Congress is from the
papers of Speaker Thomas S. Foley, Cage 655, Box 418, folder “Press
Files/1990/Democratic Strategy,” TSF. The 103rd Theme Team roster is
from the letterhead of the Theme Team in Lamar Smith “Dear Republican
Colleague,” February 4, 1994, JHQ. The 104th Theme Team roster was
located in a Martin Hoke letter dated March 14, 1996, F “Omnibus CR,”
Box 2263, NLG.
15. It should be acknowledged that many APD scholars are as much focused
on stability as they are on change.
16. Indeed, what follows is largely an analysis of several roundtables and
exchanges among APD practitioners on what the subfield is and how it
should be considered by its practitioners and the rest of the field.
17. Although there are impressive APD works that focus on one institu-
tion (Tulis 1987; Schickler 2001), APD has prized analyses that escape
traditional subfield boundaries that focus on one institution, such as the
Congress, the presidency, or the courts.
18. Defending APD against the criticism that it ignores methodological ques-
tions, Skowronek wrote, “The criticism I get from historians runs just
the other way: that APD is too intent to cut through the gristle to some
allegedly deeper truth, that our analyses are too preoccupied with estab-
lishing general propositions and illustrating overarching concepts” (2003,
107).
19. This is, of course, true of more traditional and scientific political science.
Morris Fiorina wrote of the value of history to the positive theory of
institutions: “PTI scholars are not attempting to displace other kinds of
historically oriented scholars; PTI scholars need them. PTI scholars value
social science history precisely because most PTI scholars are not going
to do the comprehensive, detailed historical research. Rather, they will
rely on historians to describe the processes and institutions and identify
relevant data” (1995, 112). The extent to which scholars such as Fiorina
who consume historical information vastly outnumber the political scien-
tists who might produce that kind of information represents a serious defi-
ciency in the discipline—and an unsustainable one which leaves political
scientists no option but to rely on the work of historians.
20. As Ian Lustick (1996) has pointed out, historians do not regard their work
this way.
Bibliography
Arnold, R Douglas. 1982. “Overtilled and Undertilled Fields in American

Politics.” Political Science Quarterly 97: 91–103.
Bates, Robert, Avner Grief, Margaret Levi, Jean-Laurent Rosenthal, and Barry
R. Weingast. 1998. Analytic Narratives. Princeton, NJ: Princeton Univer-
sity Press.
Bensel, Richard. 2003. “The Tension between American Political Develop-
ment as a Research Community and as a Disciplinary Subfield.” Studies
in American Political Development 17: 103–106.
Burnham, Walter Dean. 1982. “Party Systems and the Political Process.”
In The Current Crisis in American Politics, 92–117. New York: Oxford
University Press.
Cooper, Joseph, and David W. Brady. 1981a. “Institutional Context and Lead-
ership Style: The House from Cannon to Rayburn.” American Political
Science Review 75: 411–425.
———. 1981b. “Toward a Diachronic Analysis of Congress.” American
Political Science Review 75: 988–1006.
Fenno, Richard F. Jr. 1986. “Observation, Context, and Sequence in the Study
of Politics.” American Political Science Review 80: 3–15.
Fiorina, Morris P. 1995. “Rational Choice and the New(?) Institutionalism.”
Polity 28: 107–115.
Gamm, Gerald, and Kenneth Shepsle. 1989. “Emergence of Legislative
Institutions: Standing Committees in the House and Senate.” Legislative
Studies Quarterly 14: 39–66.
Gerring, John. 2003. “APD from a Methodological Point of View.” Studies in
American Political Development 17: 82–102.
———. 2004. “What Is a Case Study and What Is It Good For?” American
Green, Matthew N. 2010. The Speaker of the House: A Study of Leadership.
New Haven, CT: Yale University Press.
Harris, Douglas B. 1998. “The Rise of the Public Speakership.” Political

Science Quarterly 113: 193–212.
———. 2005. “House Majority Party Leaders’ Uses of Public Opinion Infor-
mation.” Congress & the Presidency 32: 133–155.
———. 2008. “Let’s Play Hardball: Congressional Partisanship in the Televi-
sion Era.” Paper prepared for presentation at “Going to Extremes: The Fate
of the Political Center in America,” a multidisciplinary conference at the
Nelson A. Rockefeller Center for Public Policy and the Social Sciences,
Dartmouth College, June 19–21.
———. 2009. “Recovering History and Discovering Data in the Archives:
An Alternative Mode of Research for Congress Scholars.” In An Amer-
ican Political Archives Reader, edited by Karen Dawley Paul, Glenn R.
Gray, and L. Rebecca Johnson Melvin, 429–440. Lanham, MD: Scarecrow
Press.
———. 2010. “Partisan Framing in Legislative Debates.” In Winning With
Words: The Origins and Impact of Framing, edited by Brian F. Schaffner
and Patrick J. Sellers, 41–58. New York: Routledge.
Hewlett, Richard. 1975. “Government History Writing from the Inside.”
In Research in the Administration of Public Policy, National Archives
Conferences, Volume 7, Papers and Proceedings of the Conference on
Research in the Administration of Public Policy, 7–11. Washington, DC:
Howard University Press.
Jacoby, Sanford. 1990. “The New Institutionalism: What Can It Learn from
the Old?” Industrial Relations 29: 316–359.
Katznelson, Ira, and John S. Lapinski. 2006. “At the Crossroads: Congress and
American Political Development.” Perspectives on Politics 4: 243–260.
Kolodny, Robin. 1998. Pursuing Majorities: Congressional Campaign
Committees in American Politics. Norman: University of Oklahoma Press.
Lustick, Ian. 1996. “History, Historiography, and Political Science: Multiple
Historical Records and the Problem of Selection Bias.” American Political
Maltese, John Anthony. 1994. Spin Control: The White House Office of
Communications and the Management of Presidential News. 2nd ed.,
revised. Chapel Hill: University of North Carolina Press.
March, James G., and Johan P. Olsen. 1984. “The New Institutionalism: Orga-
nizational Factors in Political Life.” American Political Science Review
78: 734–749.
Orren, Karen, and Stephen Skowronek. 2004. The Search for American Polit-
ical Development. New York: Cambridge University Press.
Roof, Tracy. 2011. American Labor, Congress, and the Welfare State, 1935–
2010. Baltimore: Johns Hopkins University Press.
Ross, Dorothy. 1991. The Origins of American Social Science. New York:
Cambridge University Press.
———. 1995. “The Many Lives of Institutionalism in American Social
Science.” Polity 28: 117–123.
Salisbury, Robert H., and Kenneth Shepsle. 1981. “U.S. Congressman as
Enterprise.” Legislative Studies Quarterly 6: 559–576.
Schattschneider, E. E. (1942) 1970. Party Government. Westport, CT: Green-
wood Press.
Schickler, Eric. 2001. Disjointed Pluralism: Innovation and the Development
of the U.S. Congress. Princeton, NJ: Princeton University Press.
Scott, W. Richard. 1995. Institutions and Organizations. Thousand Oaks, CA:
Sage.
Skemer, Don C. 1991. “Drifting Disciplines, Enduring Records: Political
Science and the Use of Archives.” American Archivist 54: 356–368.
Skocpol, Theda. 1984. “Emerging Agendas and Recurrent Strategies.” In
Vision and Method in Historical Sociology, edited by Theda Skocpol, 356–
391. Cambridge: Cambridge University Press.
———. 1995. “Why I Am an Historical Institutionalist.” Polity 28: 103–106.
Skowronek, Stephen. 2003. “What’s Wrong with APD?” Studies in American
Political Development 17: 107–110.
Smith, Rogers M. 2003. “Substance and Methods in APD Research.” Studies
in American Political Development 17: 111–115.
Swift, Elaine K. 2002. The Making of an American Senate: Reconstitutive
Change in Congress, 1787–1841. Ann Arbor: University of Michigan
Press.
Tulis, Jeffrey. 1987. The Rhetorical Presidency. Princeton, NJ: Princeton
University Press.
Chapter 3
Pulling�Back�the�Curtain
Archives�and�Archivists�Revealed
Linda A. Whitaker and Michael Lotstein
Those who do not themselves use the Archives are indirectly

affected by the ideas and findings of those who do. This is the
Multiplier Effect: the research done by a few political scientists,
social scientists, and historians cumulatively informs the work of
other academics, as well as government officials, journalists, and
the public at large.
–Bartholomew H. Sparrow, University of Chicago
Answering�the�Call
With the exception of an enterprising few, political scientists have been out
of the archives for the past fifty years or more (Skemer 1991). This means
that nearly three generations of political science scholars have little or no
experience with primary sources. In a parallel universe, nearly three gener-
ations of archivists have had little or no contact with political scientists.
The Frisch-Kelly call for political scientists to return to the archives is
both timely and serendipitous (Frisch and Kelly 2003, 2004, 2005, 2009).
It challenges the thinking within both disciplines and in turn promises

to invigorate the professional literature (not to mention the professionals
themselves), increase collection use, and improve reference services to
political scientists. Returning to the archives could not come at a better
time because it takes advantage of the sweeping changes in the archival
profession—most of which are intended to benefit the researcher.
This chapter attempts to bridge the fifty-year hiatus with an accelerated

reentry into the current archival environment. It is intended to provide
insights for both the researcher who is new to archives and the archivist
who may be asked to think in data sets. Though political scientists
have been largely absent from special collection reading rooms, they are
coming back. For those who remain unconvinced, our goals are to demys-
tify the archives and provide an “insiders’” guide to archival research.
This is to assure the political scientists who answer their colleagues’ call
that archivists are willing partners with a vested interest in a researcher’s
success. For a discipline such as political science which is driven to count
things, archives provide a wide range of countable opportunities.
An�Aerial�View�of�the�Archival�Landscape
The Language and Culture of Archives
Archives? What, like, Government Documents?
We have Gov Docs in our library.1
Archives have rituals, customs, and a language all their own. If one
cannot speak the language, one cannot ask the right questions. Under-
standing the rules of engagement and the basic differences between a
library and an archive are fundamental to navigating the system. This is
called doing one’s homework, and if it is done well, the gates will give
way to a brave new world.
Pulling Back the Curtain 103
Archives are largely distinguished from libraries by their rare, unique,

original material that has not been commercially published. Archives (a
physical repository or a virtual database) are where many primary sources
can be found. The Society of American Archivists (SAA) defines archives
(as well as an archive) as
materials created or received by a person, family, or organization,

public or private, in the conduct of their affairs and preserved
because of the enduring value contained in the information they
contain or as evidence of the functions and responsibilities of their
creator, especially those materials maintained using the principles
of provenance, original order, and collective control; a permanent
record. (Pearce-Moses 2009)
Primary sources are defined as
material that contains firsthand accounts of events and that was

created contemporaneous to those events or later recalled by an
eyewitness. Primary sources emphasize the lack of intermedi-
aries between the thing or events being studied and reports of
those things or events based on the belief that firsthand accounts
are more accurate. Examples of primary sources include letters
and diaries; government, church, and business records; oral histo-
ries; photographs, motion pictures, and videos; maps and land
records; and blueprints. Newspaper articles contemporaneous with
the events described are traditionally considered primary sources…
Artifacts and specimens may also be primary evidence if they are
the object of study. (Pearce-Moses 2009)
To insure that rare, unique, original (think priceless) documents are

available for the largest number of researchers for the longest period
of time, security measures such as prohibiting food and beverages and
requiring visitors to sign in, show identification, remove backpacks, and
view items under supervision in a reading room are de rigueur. The mate-
rial does not circulate, which means that nothing can be checked out. Note
that government documents, even rare and endangered print material, can
be found in open stack areas.
How archival material is described bears no resemblance to the standard

publication or book format that is found in library catalogs. The archives
call numbers are unique to the repository and reflect many items aggre-
gated together into a collection. The archival descriptions vary signifi-
cantly and depend on an archivist to create them.
Who Archivists Are and What They Do

Forget whatever dusty images come to mind when archives or archivists
are mentioned. It was never true back in the day and is light years away
from those stereotypes today. Political scientists of the 1950s would not
recognize the contemporary research environment and archival setting;
the staff and reading rooms bear little resemblance to those of the past.
The sleepy world of rare books and manuscripts has been replaced with
a beehive of activity.
There are many reasons for this, not the least of which is educational
preparation and the profession’s response to technology. In the past forty
years, the position of an archivist has leaped from that of a shop apprentice
(no degree required) to requiring graduate degrees for entry-level posi-
tions (O’Toole and Cox 2006). At the heart of this progress has been a
quest for a distinct identity, parity with other professional disciplines, and
public visibility. The very nature of the work defies a high profile. It is
by turns quiet, solitary, behind the scenes, and analytical. Archival work
can be found in print, in library and museum exhibits, online, in podcasts,
on listservs, on blogs, on YouTube, on Facebook, and on Twitter. In other
words, archivists can be found everywhere, though it may not be obvious.
Know this: What archivists do is neither a walk down memory lane nor a
serene contemplation. The work cannot be neutral.
Archivists play a decisive role in “shaping the historical record” (Blouin

1999; Cook 2001; Greene 2002; E. Kaplan 2000). What they collect, what
they discard, how they describe material, and how they connect mate-
rial, not only to researchers but to other materials, are direct responses
to the very nature of twentieth- and twenty-first-century records. What
archivists do, how they do it, and why they do it affects the meaning,
context, and interpretation of archives. They are forced to be pragmatic
and are largely driven towards use and value over universal truths or
absolute objectivity. In the process, they move, shovel, squeeze, edit, and
reject. Astute observers outside the archival profession have recognized
(and perhaps have compensated for) these facts of archival life when they
conduct research. Unfortunately, archival decisions are rarely disclosed
in a finding aid or personally witnessed by researchers (Light and Hyry
2002).
The archival enterprise shares common ground with the evolution of
political science. Like political scientists, archivists have moved away
from certain tenets of historians (Cook 2009; Skemer 1991). Over
the years, archivists looked elsewhere for methodologies and concepts
that advanced their own theories and practices. Driven by technology,
archivists have appropriated information science as “the mother disci-
pline” (Greene 2003–2004). The archival profession now fits squarely
within other applied sciences such as archeology, engineering, and infor-
mation technology. In this first decade of the twenty-first century, half of
the five thousand archivists in the United States come from library and
information science (LIS) programs.
Archivists are omnivorous and like to hunt. They are hard-wired to

search across disciplines, formats, and other repositories for answers.
They are always on the lookout for historic material held by private
individuals, agencies, and organizations—original sources that may not
appear in any archive. As political science becomes more interdiscipli-
nary, archives should be viewed as simply one more area for investigation.
Political scientists must go wherever the data lead, even if it means plan-
ning an assault on archival barricades.
You Can Run but You Cannot Hide

Like political scientists, LIS scholars and practitioners have adopted
observation and measurement tools. Studies of information-seeking
behaviors and the use of source materials as well as repository and

researcher surveys dominate the professional literature. It should come
as no surprise that political scientists appear in the library and archives
literature. Frisch and Kelly may be relieved to know that many of their
observations reported in 2009 have been independently confirmed in the
findings of this research.
As the objects of study, much is known about the source preferences,

aversions, and unsuccessful searches of political scientists, whether in
the library stacks or online. Political scientists along with economists
are among the largest single users of government documents (Hartnett
2009). Political scientists share common ground with researchers from
other disciplines: they do not like microforms, they are reluctant to ask a
librarian for assistance, and they prefer to work outside the library system
and to go it alone (Straw 1993). Many archivists understand and may
secretly agree with these sentiments, but the difference is that they would
not walk away without an answer.
Based on information-seeking behavior studies, the list of “failures to

find” for political scientists is instructional. The most frequently reported
areas of frustration are (1) out-of-print federal and congressional docu-
ments, (2) gray literature (print and electronic documents that were not
commercially published), and (3) older regional newspapers (Straw 1993).
The fact that government documents rank first in what political scientists
(arguably the connoisseurs of the Gov Docs genre) cannot find offers a
teaching moment.
The avoidance behaviors described earlier combined with the nature of

these elusive primary sources is the perfect research storm. The formula
for failure could read: researcher habit + resistance to change ÷ X time
+ energy expended = ± meaningful yield. Assuming some readers have
encountered similar frustrations, what follows is a mini boot camp on Gov
Docs. This spares the shyest of scholars from the embarrassment of having
to ask. If one is to appreciate the research potential of congressional collec-
tions, one must start here.
Listen Up
In the spirit of what Skemer (1991) described as “aggressive archival
outreach,” let us review the types of official federal publications (print
and electronic) that are found in a typical library Gov Docs department.
These sources include but are not limited to bibliographies, indexes, and
directories to all levels of government; department and agency reports,
budgets, and statistics too numerous even for political scientists to count;
congressional biographies; congressional hearings; the Congressional
Record; Congressional Research Service reports; committee prints; statis-
tical abstracts; census data; judiciary and executive branch documents; the
U.S. Code and rules; the Federal Register; patents; maps; election results;
campaign financial reports; economic data from multiple sources; and
grants. And that is why the U.S. government is the largest single publisher
in the world.
To be sure, many of these materials are notoriously difficult to find

and may be impossible to locate in one library or in a database, espe-
cially county and city publications, which often require an on-site visit.
Every state has a state library and a state archives. Arizona is in the
small minority that combines these functions, but state librarians and
state archivists will acknowledge just how difficult it is to enforce agency
compliance with depositing their records.
The thicket known as Gov Docs is an acquired taste among librarians.

Those who are drawn to it are not only knowledgeable, but they also have
a passion for the material. Most have created handouts for researchers
because the federal classification system is like none other. To add to the
muddle, new federal digitization initiatives have the effect of adding more
layers to the search without eliminating the need to rely on print material.
Gov Docs librarians find themselves in the midst of significant change.
Many of the remaining 1,250 federal depository libraries (down from

1,400) do not have their pre-1976 holdings represented in their online cata-
logs. As much as 20 percent of this material is congressional (Hartnett
2009). Of these survivors, most are “selection depositories” that carry only
a fraction of the available government material. There are one or two “full”
depositories in each state—usually academic libraries—that retain a copy
of all government publications. The exception is Wyoming, which has no
federal depository and is not served by a regional depository.
A Gov Docs librarian will know where to find these hidden sources
locally or elsewhere. If certain congressional documents cannot be found
online using the LexisNexis® Congressional database, ask about the
microfiche set produced by the Congressional Information Service from
1789 through the present. It includes an impressive number of unpublished
materials. A monthly index of federal publications dating back to 1895
is also available. If a researcher wants to chat online, Government Infor-
mation Online (http://govtinfo.org) connects users with librarians who
have “a specialized knowledge of agency information dissemination prac-
tices as well as expertise in how to use government information products,
resources and or publications…These librarians are dedicated to helping
users meet their government information needs.”
Since the 1980s, there has been a trend among libraries to discard older
Government Printing Office (GPO) physical material in favor of all things
digital. Some librarians, trying to rescue old, fragile federal publications
and transfer them to their rare books and manuscript departments, have
been rebuffed because this material is not viewed as rare or unique. Why?
Originally, these publications were widely distributed and therefore are
assumed to be ubiquitous. For lack of a better term, federal documents of
a certain age have become an endangered species.
This has significant implications for archivists who traditionally have

pulled committee prints and other GPO documents from collections and
transferred them to the Gov Docs department. The downstream effect for
researchers is no less significant. Note that GPO Access is transitioning to
the Federal Digital System, or Fdsys, in order to keep up with changes in
technology. This is touted as a more sophisticated and robust database for
finding federal publications. Does this mean that everything a researcher
might need is now digitized? Of course not. The information captured in
this new database is described as “contemporary” and dates back only to

1994. Gov Doc librarians now troll the GPO “Needs and Offers List” for
the very publications that were tossed earlier in favor of online sources.
But if a researcher is microform aversive and does not want to ask a
librarian—anything, anytime, anywhere, anyhow—the fact that a federal
document might be at the library is moot.
Here is the irony: the kinds of material that are on the political science
“failure to find” list abound in congressional collections. For govern-
ment documents, there are committee prints (published and unpublished),
debates and testimonies, party campaign and party policy committees,
hearings (open and closed), party vote analyses, and agency budgets.
For gray literature, there are federal grant applications (tribal, corporate,
academic, nonprofit) with supporting statistical analyses, corporate tech-
nical reports, research data, policy impact studies as background for testi-
monies in legislative hearings, patent applications with technical reports,
white papers, memos, economic forecasts, budget analyses, and topical
reports submitted by think tanks, task forces, government contractors, the
Department of Defense, and the GAO. For regional newspapers, there are
news clippings provided by a daily professional clipping service (local,
regional, and national), mounted in scrapbooks and continuously main-
tained by staff for the duration of the officeholder’s career—standard oper-
ating procedure for all members of Congress. In the Goldwater papers
alone, there are 195 scrapbooks of news clippings from 1935 to 1985,
all digitized to searchable CDs. This is low-hanging fruit that is ripe for
picking.
Harris, Kelly, Frisch, Parker, and a growing number of investigators

may not see congressional papers as exactly easy pickings, but they do
agree on their research value. They want metrics, longitudinal analyses,
and a fresh way to test theories. Congressional collections meet or exceed
those criteria. They include a wide variety of quantitative documents and
in great numbers—archivists just have not described or viewed them that
way. The records usually span a sufficient number of years to conduct
longitudinal studies. Congressional newcomers start as a “class” and can

be viewed as a cohort until they leave office. How did they behave as a
class over time? Who rises to the top and who gets left behind? Leader-
ship studies, political ecology, corporate political activity, and institutional
studies are fascinating and critical to the understanding of complex inter-
relationships. Archivists do not have to understand the statistical compu-
tations that are applied to research problems in order to appreciate the
fresh perspectives of the findings.
Most of the documents described here are largely untapped, are often
ignored by historians, and have rarely—if ever—been used to create data
sets. What does all of this mean to the political scientist? The time has
never been better for doing archival research.
Modern�Collections,�Collecting,�and�Collection
Management
Dilemmas of the Twentieth and Twenty-First Centuries

By now, the savvy reader should be asking, If the material is so rich and
the timing is so good, why are not more political scientists using it? This
has less to do with pushing the research norms of political science than one
may think. Even the most adventurous, inspired, and motivated find the
going rough (Harris 2009; see also chapter 4 in this volume). Their obser-
vations, frustrations, and cautions are the very same that archivists have
voiced for the past forty years: bulk (way too much), redundancy (dupli-
cates of duplicates), missing data (not created, not collected), imperma-
nence (never meant to last), and bias (overdocumentation in one area to
the exclusion of another) (Ham 1984). This is the very essence of twen-
tieth-century collections and will likely hold true for those in the twenty-
first century. Never has so much been collected and finding useful infor-
mation been so difficult. Gerald Ham described this pandemic as “the
noise and distortions of a badly tuned television set” (1984, 12). Today,
the same could be said of the Internet and social networking.
The issues associated with contemporary collections and congressional

collections in particular are well documented in the archival literature.
Congressional collections were identified early on as significant contribu-
tors to the avalanche of material that threatened to bury the archival enter-
prise and to drain repository budgets. The staggering national backlog has
stirred heated debates about ethics, access, and collecting practices. It was
called an archival crisis then. It is an archival crisis now. Many congres-
sional collections languish in repository backlogs, hidden and relatively
unknown to archivists and researchers alike. How big are these backlogs?
They are so big that the Bieneke Library at Yale closed down for a
year and did nothing but process its backlog. They are so big that Mark
Greene at the American Heritage Center at the University of Wyoming
was driven to develop a revolutionary but disciplined approach to eman-
cipate the best material from the backlog and eliminate the rest (Greene
and Meissner 2005). They are so big that some backlogs equal or exceed
processed collections. Given the magnitude of the task and the general
lack of resources to accomplish it, none of this should come as a surprise
except for one thing: it has galvanized the archival profession—not known
for moving at a hurried pace—into a rush for solutions.2
Invariably, it falls to the newest archivist with the least experience to

process congressional collections. Many vow never to do it again. Further,
these collections contain every size and type of paper product and format
known: audiovisual tapes and reels, film, albums, vinyl records, news-
papers, photographs, audio cassettes, dictation belts, key punch cards,
memorabilia, and artifacts, to name a few. This makes them difficult to
process, difficult to describe in traditional library catalogs, and expensive
to store and also calls for special preservation techniques that are beyond
the scope of most repositories.
Congressional collections are the poster children for all of the archival
sins and excesses of the modern age. These are the “bad boys” of
the archives world. They are notorious for their bulk and complexity.
Congressional collections span gender, race, party affiliation, geograph-
ical distribution, and media format. The records are at once local, regional,
and national in scope. They are ubiquitous yet unique. They can be
found in repositories large and small and in every corner of the country.
Frisch and Kelly described this diaspora as “not exactly tourist destina-
tions” (2005, 1). Understanding how these collections got that way will
help to explain the inherent unevenness, omissions, messiness, and unpre-
dictability from one collection to the next.
These characteristics, endemic to congressional collections, appear to

concern political scientists (Harris 2009). However, it should not be the
limiting step in using archival material. The topography of a collec-
tion may slow navigation, the materials may require interpretation, and
the arrangement and description may inadvertently introduce bias in
selection. Once these confounding factors are identified, analyzed, and
reported, scientists in all disciplines internalize, accommodate, and move
on.
The top twenty senior U.S. senators have been in office for twenty-three
to fifty-one years. The records generated by a typical senator’s office have
been estimated at one hundred linear feet or one hundred large boxes annu-
ally (Aronsson 1984). Each large box holds approximately 2,200 docu-
ments. Based on the 1984 calculations, the number of boxes (2,300 to
5,100) and documents (5,060,000 to 11,220,000) that could be found in the
senior senators’ collections is staggering—even for archivists who prefer
big puzzles. From all anecdotal reports, the records generated per year
since 1984 have increased despite the advent of electronic records. For the
political scientist, that is all the more to count.
Along with the alarming rate of backlogs, doubt was cast on the research
value of congressional collections when they were measured against the
costs to process them. Additionally, there was a steady decline in their
use (Aronsson 1984). The approximately two hundred U.S. archivists who
work with this material must make the case that these collections should
be a priority for processing and that they should be actively collected when
the resources are there to support them. Critical to the success of making
these collections sustainable is an appraisal by an archivist before a collec-

tion is donated and transferred. Many trials and tribulations of reposito-
ries, donors, and archivists could be avoided with this one initial step.
Appraisal is “the process of identifying materials offered to an archive

that has sufficient value to be accessioned” (Pearce-Moses 2009). For
reasons that are discussed later, archivists are often not involved when
institutions are courting congressional papers. Be aware: archivists do not
save everything, particularly in congressional collections, but neither do
they throw things away indiscriminately. There are standards and best
practices established just for these collections. Given the expense for
storage, preservation, and access, some administrators see these collec-
tions as underutilized and not worth the expense. Some repositories
(wisely) will no longer accept congressional collections, recognizing their
incapacity to support them. But for those that do, the opportunities are
nearly endless.
The life cycle of these collections, from creation to citation, impact

what can be found in the descriptions online and in the folders in the
reading room. The gulf between the computation of data sets (relatively
inexpensive and efficient) and the uncertain terrain of twentieth-century
records (comparatively expensive and inefficient) can be disorienting,
cause doubt, and make for difficult choices in the field (Skemer 1991). All
of these points are true, but such is the nature of scientific investigation.
In the end, the lure of discovering something totally new and unexpected
may prove irresistible. What follows is a “natural history” of the typical
congressional collection.
How Congressional Collections Are Created

It is important to understand that congressional collections are hybrids
in the archives world—they are neither strictly governmental records nor
strictly personal papers (Aronsson 1984). Also, it should not be forgotten
that they are generated and stored at taxpayer expense. Most of the docu-
ments that are found in these collections were created as a result of the
functions of the elected official and his or her office staff. They include,
but are not limited to, congressional committees and caucuses, committees
of political parties, commissions, legislative agendas, bills, constituent
service, staff projects, and campaigns. Functions of the individual/posi-
tion/office are the building blocks for how these collections are arranged.
Official committee records are preserved by their respective houses in
Congress.
In 2008, H. Con. Res. 307 formally recognized the historical impor-

tance and research value of congressional members’ papers. For the first
time, members were charged to “take all necessary measures to manage
and preserve” their papers, which “serve as indispensable sources for the
study of American representative democracy.” Senators and congressmen,
therefore, are officially encouraged but not legally bound to donate their
papers. Many give them to their alma maters or various state institutions,
which explains their wide and often remote distribution.
Up until the first Senate office handbook was published in 1985,

there were no records management guidelines. This handbook identified
records that were temporary or permanent. The guidelines were volun-
tary then and remain voluntary today. The elected official and his or
her office staff determine how they manage their records and files. The
disparity from one congressional collection to the next starts from the
day the first documents are generated and filed. These differences often
reflect the personalities creating the documents. Those scholars who are
familiar with Senator Harrison Schmitt’s office files say he saved every-
thing. Senator Dennis DeConcini, in contrast, did not. His office files,
though they are spare, concentrate on the essentials: staff meeting minutes,
staff lists and assignments, and office protocols and procedures. The files
of his legislative directors, legislative aides, and administrative assistants
are remarkably intact and include background material, analysis, memos,
day-to-day contacts, conversations, and reports on special assignments.
In Senator Barry Goldwater’s office files (administrative series), there are
few internal staff communications, but the volume of material on federal

nominations is significant.
Attempts to influence and standardize office filing systems continue
to this day. Guidelines for opening and closing Senate offices have
been developed. The Senate office records management handbook was
revised in 2003. It was not until 2006 that a similar manual was created
for the House of Representatives. Most congressional offices have now
adopted more systematic practices for managing records. In the near term,
archivists outside the beltway know they will be receiving congressional
collections that bear little resemblance to established records management
protocols. This means that researchers and archivists will continue to find
these collections uneven and unpredictable.
Retention schedules such as those outlined in the House and Senate

handbooks inform the office staff of what to keep and what to throw
away. The guidelines were developed by Senate and House archivists who
sought to protect material with research value. The concept of research
value requires a sidebar because it directly affects political scientists and
may explain what is and is not in a collection.3
Enduring research value is a frequently used term and a recurring topic

of discussion in archival literature. In the lexicon of archivists, research
value is tied to current and future use among other criteria. According to
Elsie T. Freeman, it is also in the eye of the beholder. She argued, “Records
are inert until they are acted upon by the human mind…records do not
exist until they are used” (1984, 118). Research value, therefore, is not the
same as historical value. Records, especially twentieth- and twenty-first-
century documents, are not museum artifacts. Research value has often
been defined by user studies, information-seeking analyses, and surveys
of historians. As a group, historians feature prominently in the archival
literature, suggesting they are the largest and perhaps the most significant
users. A search using historian as a keyword reveals that 1,732 articles
about or relating to historians have appeared in the American Archivist
since it was first published in 1938. What about political scientists? For
them, there is but one article, and that is Skemer’s persuasive argument to
think beyond the reference needs of historians.
Are historians the biggest consumers of archival material? They
certainly use collections intensely, but do they outnumber everyone else?
Based on our observations, the answer is no. Urban planners, political
scientists, geographers, environmentalists, architects, archeologists, engi-
neers, lawyers, journalists, geologists, educators, government officials,
authors, film documentarians, social activists, photographers, designers,
genealogists, high school and college students, campus visitors, and
alumni outnumber historians in aggregate and on any given day. Archives
are not, may never have been, and should never be the domain of one
discipline (Brooks 1951; Skemer 1991).
The ramparts were breached long ago. The democratization of archives

has occurred over time, which provides ballast for the archival enterprise
and unlimited opportunities for archivists. This bodes well because none
of the researchers listed previously use any archival material to create data
sets that measure the problems or questions that are generally studied by
political science.
The records of congressional members that do survive the retention

schedules are usually stored in Suitland, Maryland. The boxes are ware-
housed free of charge for up to one year after the officeholder leaves
Capitol Hill. For some, the disposition of their papers is methodical and
deliberate, with a repository identified well in advance (e.g., Senator Orrin
Hatch). For others, especially those who unexpectedly lose an election,
die in office, or suddenly decide not to run, the exodus can be swift and
disorganized (e.g., Senator Dennis DeConcini). Those who seek reelection
or have other political aspirations often keep their papers, which end up
in basements, garages, attics, or warehouses. Often the boxes are moved
multiple times. When congressional papers arrive at their final archival
destination, they may no longer resemble how they looked on Capitol Hill.
Much of the Goldwater material had stops along the way.
The Acquisition Dance

How do congressional collections come to a repository? The answer is, not
easily and often not in a straight path. First and foremost, these collections
can generate a political climate all their own. They tend to be high stakes,
high profile, and come with even higher donor expectations. Generally
speaking, the higher the political profile, the more repositories throw
their hats into the ring. The higher the stakes, the more promises are
made. When a highly valuable collection becomes available, there will be
competing interests. And so it was in 2008, when the fate of Senator Pete
Domenici’s papers was played out in the media.
Competition and controversy arose when two of New Mexico’s largest
universities vied for the privilege of housing the Domenici collection.
In the end, his alma mater lost to the rival institution, which offered
an impressive package: a Domenici Archives in the campus library;
the Domenici Legacy on Public Policy Conference, held annually and
featuring prominent speakers; the construction of a building and garden to
honor the senator and his wife; the Domenici Institute for Public Policy;
and the publication of a book on Domenici’s career. This required a
financial commitment that few public institutions could sustain. In this
case, a blue ribbon fund-raising committee was created to support various
elements of the agreement. Fund-raising is ongoing.
Few of these negotiation points or peripheral agreements are reflected in

the deed of gift that the donor and repository representatives sign. A deed
of gift is “an agreement transferring title to property without an exchange
of monetary compensation” (Pearce-Moses 2009). This is a legal, contrac-
tual agreement between the donor and the repository or institution. It
establishes the conditions governing the transfer of title and sets out any
restrictions or conditions for use. In academic institutions, the signatory
of this legal instrument could be the dean of libraries and/or the head of
special collections but is frequently the university president. For a high-
profile congressional collection, the archivist may not be consulted, espe-
cially if he or she is new in the position or was hired as a project archivist,
a position which is often temporary and on soft money. Agreements with

the institution apart from the deposition of papers may not be known or
fully understood by those who are charged with managing the archival
materials.
Nothing challenges public relations for an institution more than an

unhappy donor who also happens to be a former congressman. A recurring
issue is the time it takes to make the materials available. Many deeds of
gift come with a vaguely worded phrase such as “the repository agrees to
make this collection available in a timely manner.” If funding is not forth-
coming, collections may be processed intermittently as time and money
allow or may sit untouched, sometimes for years. Library directors who
inherit these collections may see this as an unfunded mandate. Complaints
from well-placed donors to an institution’s president or governing body
are not unknown and can have lasting effects. Donors have threatened to
remove papers, citing the failure of the repository to fulfill the obligations
stated in the deed of gift. Sometimes, the family takes matters into its own
hands.4
More institutions are asking for funds to support these collections

before they arrive. The papers of Senator Ted Stevens at the University of
Alaska, Fairbanks (UAF) underscore the size, costs, and compromises that
come with congressional collections. British Petroleum donated $500,000
to fund their processing.
We were pretty upfront that we were going to need some money

because of the sheer mass of this. When we say [there are] 4,500
boxes, I don’t think people realize how big that is. UAF didn’t
have the funds or work force to deal with such an enormous
archive. Without the grant from the oil company it would have
taken decades to sort through the archive. (Development Officer
Naomi Horne, Rasmuson Library, as cited in Richardson 2009)
Of interest to researchers and outside observers is the fact that the orig-
inal documents and materials still belong to Senator Stevens or his heirs.
According to the agreement, the collection may be transferred after ten

years to a university library in Anchorage if a facility there can house it.
The collecting environment described here is not for the faint of
heart, but it does typify the background of many congressional collec-
tions. These scenarios are so commonplace that they inspired Managing
Congressional Collections by Cynthia Pease Miller (2008). This manual
was written for three audiences—the officeholder, the repository, and the
archivist—all of whom have obligations of stewardship. Although it is
too soon to measure the impact, this publication establishes best practices
for all three parties. Ultimately, all of these factors—access, the condition
of the materials, the ability to navigate the bulk, making sense of gaps in
the records, locating unprocessed collections, and finding someone with
enough institutional memory to answer questions—converge to affect the
researcher.
Figure 3. Archivist Mary Anne Hamblen in the storage warehouse with the
Ted Stevens Papers Collection (2010).
Source. Marjorie Thompson (photographer).

Processing, Arrangement, and Description

Processing is “the arrangement, description, and housing of archival mate-
rials for storage and use by patrons” (Pearce-Moses 2009). For a congres-
sional collection conceived outside of a records management program, a
radical makeover may be needed to render it into something useful. Under
the best of circumstances, it is messy, time-consuming, and expensive
work. The level of processing varies considerably across collections and
repositories. This contributes to the wide disparity and unevenness that are
reported by researchers. Processing times reported in the literature vary
significantly depending on the type of collection. From the mid-1970s to
the mid-1980s, articles about processing times and costs dominated the
journals.
In 1985, Terry Abraham at the University of Idaho conducted a study

attempting to establish baseline processing rates and to calculate the
time between the arrival of material and when it was made available to
researchers. This project was inspired by the introduction of word proces-
sors, which had greatly streamlined creating the finding aid. It was called
a watershed in preparing a collection for research use. Abraham found that
photograph and oral history collections (congressional collections have
both formats) took forty-seven hours to process one cubic foot and on
average remained on the shelf for thirteen months before the processing
began. Manuscript collections (fairly uniform typescript material) took
thirty-six hours per cubic foot and on average stayed idle on the shelf
for eighteen months. For the purposes of this study, a large collection
was defined as greater than one cubic foot—microscopic compared to the
congressional goliaths.5 The bottom line is that even with the advent of
word processors, the business of traditional processing was and is glacial.
Very little about processing is transparent. It generally occurs behind

closed doors. Occasionally the collections are removed to an off-site
facility that has adequate space. Rarely are these collections processed in
full view of the patrons in the reading room. Once the fanfare announcing
the donation of papers dies down, the material typically sinks out of sight
until it emerges some years later as a collection that is open and available
for research. Processing is rarely reported while it is work in progress.
Archivists working with the Goldwater papers broke with tradition. The
Goldwater progress notes were posted on the web over a five-year period.
They were targeted at several audiences: the Arizona state legislature, who
allocated $529,000 to insure the long-term preservation and access to the
collection; the family, friends, and former staff members who had voiced
many concerns through the years; researchers who were curious about
access; a board of directors who were unfamiliar with archival procedures;
and colleagues who were interested in demystifying what archivists do.
Due to limited space and staff, processing migrated into the reading
room. There were risks, criticisms, and warnings about this move; much
of the concern had to do with security breaches, theft, or interference
from patrons—none of which happened. The benefits (teaching moments,
shared discoveries, researchers’ questions, observations of how the mate-
rial was used) far outweighed the risks. Transparency proved to be a grand
experiment. A narrative and images on the Internet plus a physical pres-
ence in the reading room demonstrated how an archivist tactically thinks
about disorder, physically wrestles with it, and finally wraps it into a
package that meets archival arrangement, access, and preservation stan-
dards. These before and after images speak for themselves.
Figure 4. Personal and political papers of Senator Barry M. Goldwater before

processing (2003).
Source. Arizona Historical Foundation, Hayden Library, Arizona State Univer-

sity, Tempe, Arizona.
Figure 5. Personal and political papers of Senator Barry M. Goldwater after

processing (2009).
Source. Rebekah Tabah (photographer).
Proposed by Greene and Meissner in 2005, “more product, less

process” (MPLP) is a pragmatic response to backlogs that makes collec-
tions available to researchers more quickly. Among other things, this
means not weeding duplicates, not removing metal clips or fasteners,
having minimal arrangement, and using the original folders and boxes.
The time and money spent per box is dramatically reduced and the collec-
tions move far more quickly into researchers’ hands. The trade-off for
this accessibility is that the description of materials may also be minimal,
resulting in more time spent locating documents. MPLP has captured the
attention of (and, some people fear, polarized) the entire archival commu-
nity. Many repositories have adopted or modified the technique and report
favorable results. For unfunded collections, which represent the majority

of repository holdings, this may become standard practice.
There is a trend among repositories to make unprocessed collections
known and available to researchers. Some repositories list these online.
As with MPLP, there is a trade-off for gaining access to new, unprocessed
material. These collections are often chaotic and may not have basic
inventories, and their contents may be largely unknown. They require
much more time to navigate. Reluctant archivists, concerned about sensi-
tive material and privacy, may be more willing to allow access if the
researcher offers to make an inventory of an unexamined box. A little
cross-training and exchange of skills can go a long way towards forging
partnerships.
Arrangement is “the process of organizing materials with respect to

their provenance and original order, to protect their context and to achieve
physical or intellectual control over the materials” (Pearce-Moses 2009).
A series is the fundamental building block in arranging a collection.
Think of series as chapters that represent the functions and the formats
of the records that are found within a collection. Formats refer to specific
types of material, such as news clippings, photographs, memorabilia,
film, videos, audio recordings, CDs, DVDs, and so on. As an aside,
the media series in congressional collections can be a persuasive and
powerful adjunct for a researcher. But, as David Parker notes in chapter
4, many formats and the equipment to use them are plagued with extinc-
tion. Function refers to the purpose the documents served for the creator
of the collection (usually the donor), such as committee files or campaign
records.
Series provide the internal structure for the arrangement and description
of the collection. Series dictate how the finding aid looks and is the key to
navigation. Series can be used as outlines to locate where certain kinds of
documents are found. This becomes critical in large, complex collections
such as congressional collections. Basic series for congressional papers
include but may not be limited to personal, legislative, constituent service,
administrative, media, and artifacts/memorabilia. This is discussed further

in the “Anatomy of a Finding Aid” section.
Description is
the process of creating a finding aid or other access tools that allow
individuals to browse a surrogate of the collection to facilitate
access and improve security by creating a record of the collection
and by minimizing the amount of handling of the original materials.
(Pearce-Moses 2009)
This means the researcher must navigate a document (often hundreds of

pages in length) that is representative of the collection itself. Entire collec-
tions are not pulled for casual browsing. Original materials in their twen-
tieth-century fragile formats do not lend themselves to constant handling.
Note that the quality and quantity of descriptions vary considerably from
finding aid to finding aid and from one repository to another. The older the
collection, the less likely it will adhere to the more standardized descrip-
tions of today. Finding aids share certain features. Knowing what to look
for and how to ask the questions will spare the researcher unnecessary
expense and frustration.
Locating archival materials and making sense of them is a universe apart

from data mining online polls or election results. The differences between
what a researcher sees online or in a library catalog and the quality or
quantity of material he or she needs are critical. Studies have shown that
researchers often fail to make the association between a finding aid in
the reading room and how it is represented in a library catalog record or
online encoded in special software (Yakel and Torres 2003). To make a
successful leap from cyberspace to reality, the wise researcher will contact
the repository and ask pointed questions.6
Anatomy�of�a�Finding�Aid
A finding aid is “a tool that facilitates discovery of information within a
collection of records” (Pearce-Moses 2009). It is the most important docu-
ment in a search and will help researchers sharpen those questions. Will
the collection serve a researcher’s needs? Is a trip warranted? The finding
aid, unique to the archival profession, comes in many guises. Without it,
a collection is nearly invisible.
Finding aids were created before there were library catalogs and were
largely kept in special collections reading rooms. Many of these can now
be found on repository websites, in library catalogs, and within topical
collection guides. As a result of multirepository collaborations, there are
state and regional listings of archival collections, such as the Online
Archive of California, the Arizona Archives Online, and the Rocky Moun-
tain Online Archive.
Finding aids are comprised of descriptive elements known as “front

matter,” which introduces the collection to the researcher. It has been
said that archivists write finding aids for each other, not for the patron
(Freeman 1984). This may be true, but there are few if any alternative
proposals on the horizon. For better or worse, think of a finding aid as the
archivist’s thesis.
Collection-Level Description
A finding aid will typically begin with a collection-level description that
includes the title of the collection, the date range, the call number, and the
physical description (the number of boxes within the collection and the
linear or cubic footage). At a glance, this gives the researcher a good idea
regarding the size of the collection. These collections can span hundreds
of boxes extending many linear feet. In most cases, congressional collec-
tions are the largest collection of records held by an archive. The following
sections of the finding aid contain statements relating to provenance,
copyright, and access restrictions.
Provenance
Provenance is “the origin or source of something; information regarding
the origins, custody, and ownership of an item or collection” (Pearce-
Moses 2009). This refers to who owned the collection prior to its donation.
The term also represents a fundamental principle that the records or docu-
ments in one collection are never mixed with the records of another collec-
tion in order to insure context and authenticity. This allows the researcher
to better understand the document itself and maintains the research value
of the collection as a whole.
The provenance statement (usually very brief) describes how, when,

and by whom the collection was donated to the repository. This explains
the legal transfer (the deed of gift) of physical property from its creator
to the archives. Deeds of gift will have a clause stating that the donor
has the legal right to give the collection to an archive. This protects
the archive from fraud or legal action. The provenance statement also
alerts the researcher to any potential problems within the collection and
may explain gaps in dates, incomplete files, and missing materials. The
phrase “deposits made serially” could indicate that these materials were
not processed or cataloged at the same time or were processed years after
the initial donation. If the collection comes directly from its creator(s)
to the repository, the likelihood of loss, theft, or damage is significantly
diminished.
This raises the question, How does one know if a collection has been
interfered with (apart from the manipulations that are incurred during
archival work)? Skeptics often ask if a congressional collection has been
purged of incriminating or controversial documents by the donor, his
staff, or the archivists themselves. It certainly happens and has happened
at the highest levels of government. But is the practice common? This
is doubtful given the vast amounts of documents that are generated
during a congressperson’s tenure. Unless documents are reviewed daily
for compromising content, removed, and destroyed, a serious purge would
have to occur sometime before the shipment to a repository. At this stage,
the office staff is looking for jobs and the bulk of records may be in
transit or already stored in Suitland, Maryland. This makes an item-by-
item inspection of documents highly unlikely. Even deleted electronic
records can be recovered from hard drives.
The lament regarding modern congressional collections is that they

are comparatively “sterile” and devoid of personal information (Lucas
1978). This has less to do with purging and more to do with how records
are generated, used, and filed. Documents of a personal nature (family
correspondence, legal matters) usually do not show up in a congressional
collection. The personal series in congressional collections is intended to
capture biographical information, campaign materials, and personal VIP
correspondence—pre– and post–political career records.
A notable exception to twentieth- and twenty-first-century “sterility”

can be found in the personal and political papers of Senator Barry
M. Goldwater. The material in the personal series, as described in the
introductory scope and content note, reflects his “personal interests
and pursuits, life-long relationships, family business, public and private
service, achievements, private musings and observations, and his deep
knowledge and love of Arizona.” This series is both rich and voluminous.
Copyright
Copyright is
a property right that protects the interests of authors or other

creators of works in tangible media (or the individual or organi-
zation to whom copyright has been assigned) by giving them the
ability to control the reproduction, publication, adaptation, exhibi-
tion, or performance of their works.(Pearce-Moses 2009)
Copyright applies to intellectual, not physical, property. A repository may

own a collection of papers or photographs, but the author or photographer
may retain copyright. This means that the researcher may view the mate-
rial and take notes but may not be able use a document or image in a publi-
cation. Note that posting to the Internet constitutes a publication.
The circumstances of ownership for a body of records are explained in
the copyright statement. It is never a guarantee that a donor will transfer
copyright to a repository. If the archival institution does not own the
copyright to a collection, it is incumbent upon researchers to contact
the copyright owner to secure permission to publish, display, or redis-
tribute information from that collection. Although the copyright holder is
afforded certain rights under U.S. law, there are limitations to those rights.
Section 107 of federal copyright law governs “fair use” of materials for
the purposes of criticism, comment, teaching, and scholarship. Fair use
depends on four specific criteria:
1. The use is for nonprofit educational purposes.
2. The nature of the copyrighted work is not infringed upon.
3. Only a certain percentage of the material is used in relation to the

whole body of materials.
4. Use will not unreasonably impact the potential market for this mate-
rial in the future.
In terms of congressional collections, any publication produced by the

Government Printing Office (GPO) is considered public domain under
section 105 of federal copyright law. The researcher should look for a GPO
attribution within a document to verify its status. Note that state or local
government publications may contain copyright stipulations. The provi-
sions of section 105 also apply to certain types of unpublished documents
within a congressional collection. Any documents that are created by a
member of Congress or his or her office as a part of official duties are
in the public domain. Any personal papers or unofficial documents, such
as campaign records and personal memorabilia, may be under copyright,
depending on the donor agreement. Note that congressional collections
also contain massive amounts of copyrighted material that are sent to the
officeholder.
Restrictions
The access restriction statement will explain any prearranged restrictions
to an archival collection. Archival institutions will generally do their
utmost to ensure that no access restrictions are imposed on any collections
in their custody. The concept of “equality of access” has been a hallmark
of modern archival practice in the United States. Archivists are some-
times bound by physical and legal restrictions that require them to deny or
restrict access to collections. Occasionally, archivists will invoke restric-
tions to protect confidentiality or individual privacy.
Physical restrictions are often provided for materials that are extremely
fragile or sensitive to light and/or temperature and must be housed under
strict climatic and luminescent conditions to ensure their physical safety.
For example, a statement of restriction from the finding aid for the papers
of Congressman Paul N. McCloskey at the Hoover Institution reads:
Access Boxes 611–651 closed. Access to audiovisual materials

requires at least two weeks advance notice. Audiovisual mate-
rials may include sound recordings, video recordings, and motion
picture film. Hoover staff will determine whether use copies of the
materials requested can be made available. Some materials may not
be accessible even with advance notice. Please contact the Hoover
Institution Archives Audiovisual Specialist for further information.
Legal restrictions may come from the donor of the collection. Arrange-
ments of this nature may potentially restrict access to portions of a collec-
tion during the donor’s lifetime or after a proscribed period following the
donor’s death. This often occurs to protect the privacy of the donor or
other figures related to the collection. Archivists work very closely with
donors to craft the language of these types of restrictions as a part of the
deed of gift. Current archival practice calls for the inclusion of a sunset
date for any negotiated restrictions in order to ensure access to archival
materials in the future. An example of a statement of restriction comes

from the finding aid to the Thomas Gayle Morris congressional papers at
New Mexico State University Libraries:
Access and use of files containing personal information is restricted

to protect the privacy of individuals still alive. Users agreeing not
to reveal personal information may use these materials for statis-
tical research and reporting provided no individually identifiable
information will be disclosed.
Another variation can be seen in the Goldwater papers:
This collection is unrestricted except for certain family corre-

spondence which is closed until 2023. Although considered sensi-
tive, case files (48 boxes) have been retained and are available to
researchers. Contact the archivists or librarians for conditions and
permissions of use.
For congressional collections, restrictions may be invoked by federal

statute or national security. U.S. Senate committee files are restricted for
twenty years; House of Representative committee files are restricted for
thirty years. Senate and House restrictions imposed on official committee
records do not legally bind repositories that are in possession of personal
copies of committee material. In the Senator Dennis DeConcini papers,
restricted materials concerning nomination files for the Supreme Court are
closed for fifty years, or until January 2044.
Archival institutions should have policies in place for the researcher to

petition a review and reconsideration of access restrictions based on his
or her research needs. Repositories are well aware of the negative percep-
tions that restrictions invoke. The belief that someone is trying to hide
something can persist long after restrictions are lifted. Most public insti-
tutions will no longer accept collections with onerous restrictions.
Deeds of gift can be renegotiated, especially if the archivist makes a

case with the donor. Unbeknownst to them, two editors of this book were
the beneficiaries of such an intervention. The original deed of gift signed
by Senator DeConcini restricted access to his papers until the collection

was processed—years off at the time. The exception was his biographer,
who was allowed access to all material. Scott Frisch and Sean Kelly called
to ask about committee assignments. The information they wanted was
there. The archivists contacted the senator, explaining the ethics of equal
access and the nature of the research. Would he change the restrictions? He
agreed and signed off on the change in less than a week. This brief period
is significant considering that the original deed of gift took nine months to
negotiate. Frisch and Kelly got their documents and were none the wiser.
Moreover, it paved the way for other scholars, who would no longer have
to delay their research until the collection was fully processed.7
Biographical/Historical Note
This section is a brief summary of the functions of an organization or a
biography of an individual to document the functions, activities, events,
and changes that are essential to understanding the records. It is neither a
definitive narrative nor an interpretative work. It is intended to serve as a
chronological exposition of either the formal workings of an organization
or the highlights of an individual’s life. A separate chronology may also be
included and is very helpful in orienting the researcher to time and place
at any point in the collection.
Scope and Content Note

The more interpretative elements of a finding aid are contained in the
section known as the scope and content note. This is a summary of the
subjects, formats, conditions of materials, and the special features of the
collection and arrangement. The scope and content note should distinguish
whether the arrangement of the collection reflects original order or if order
has been imposed on it. It also should disclose information that explains
weeding, sampling, gaps in the record, and circumstances that influenced
the arrangement and description—anything that may affect how the user
plans his or her research strategies.
Due to its high research value, controversies over access and use, dete-
riorating condition, and multiple attempts at processing, the scope and
content note for the Goldwater papers lists seventeen specific archival
interventions that were taken for preservation, arrangement, description,
and access. This stripped the archival decision-making process bare, first
in a background narrative, then in a bulleted list of findings during the
appraisal. Researchers should pay attention to scope and content notes
because they summarize what can be found and what likely is not there.
Think of it as an abstract for the kinds of documents, date ranges, topics,
highlights, and unusual features of a collection. Scope and content notes
prepare a researcher’s mind for the territory ahead and ideally set up a
visual readiness to recognize the material as he or she moves through
folders.
Series
If the finding aid is the archivist’s thesis, then series are chapters that give
the collection its structure. Series are unique sets or general headings that
provide a road map for the researcher. This is an archival technique that
is intended to impose order, improve discovery, and aid in overall storage
and retrieval. At a glance, series in a congressional collection should
inform the user where to find specific information, such as photographs,
correspondence, committee files, and campaign materials, without having
to scroll through the entire finding aid.
Series represent either the function of the records or the historical

context of the records as they were created. Series may also be artifi-
cial in nature—a logical system created by the archivist to impose order
on a collection that had no discernible or systematic order. Series may
also reflect media formats such as microfilm, photographs, and memora-
bilia. The number of storage boxes within each series may also be noted,
followed by a description of the series’ contents. The following is an
example from the papers of Congressman Eldon Rudd at the Arizona State
University Libraries:
The Eldon Rudd Papers contain correspondence, transcripts, news

releases, speeches, articles, and printed matter which range in date
from 1962 to 1987. The bulk of the material dates from 1977 to
1986 and constitutes the written communications and study mate-
rials collected during Eldon Rudd’s tenure in the United States
House of Representatives. It is arranged in six series: Correspon-
dence, Publicity and Public Presentations, Legislation Files, Case
Files, Resource and Subject Files and Issues Correspondence.
Often, descriptions of significant items within the series are provided to

give the user a greater sense of what the series contains. What an archivist
will identify as significant may be based on advice and guidance from
the donor or on research performed by the archivist on other primary and
secondary sources that are relevant to the contents of the collection. For
large archival collections, quality series-level descriptions can be a valu-
able resource in narrowing the parameters for a search within a finding
aid. An example of a series-level description comes from the papers of
Congressman Eldon Rudd at the Arizona State University Libraries:
The Correspondence Series dates from 1977 to 1986 and is

arranged in four subseries: Legislative Correspondence, Depart-
ments and Agencies, Organizations and Groups, and Trip Files.
The Legislative Correspondence subseries comprise general and
congressional committee correspondence from the years 1977 to
1979. Departments and Agencies contain correspondence from
1977 to 1980 regarding the legislative concerns of various govern-
mental organizations. A selection of these folders pertains to the
movement to require metrication of weights and measures in the
United States. These are box 51, folders 13 to 13c; box 54, folders
9 to 12; box 56, folders 1 to 11; and, box 57, folders 1 and 2.
Organizations and Groups include correspondence regarding the
legislative concerns of private organizations during the years 1977
to 1980. The Trip Files subseries contain correspondence regarding
legislative proposals which were discussed at conferences outside
of Washington, D.C. during the period 1977 to 1986.
Series: Constituent Service

There is no other way to say this. Constituent service is the bane of a polit-
ical papers archivist. It is not unusual for a repository to receive constituent
mail in sacks, jammed in boxes, or just scattered loose among the more
ordered material in folders. Depending on the topic and volume, congres-
sional offices may not be able to collect all of the mail in one place, let
alone answer it.
A number of influential archivists have appraised the research value of

constituent mail and found it wanting. The reasons given were repetition,
bulk, and homogeneity, such as the use of postcards stamped with iden-
tical messages, preprinted letters sent in batches that are indistinguish-
able from each other, and well-orchestrated campaigns from groups that
apart from the first letter have little informational value (Aronsson 1984;
McKay 1978). It has been observed that opinion letters differ very little
except in volume from letters to the editors of state and local newspapers.
Congressional offices maintain statistics on constituent opinion, so why
would a researcher need to examine the bulk? Further, the opinions that
are stated in constituent mail are often at odds with local polling data on
the same topics. Lastly, voluminous constituent correspondence occupies
and competes for prime shelf space at the expense of other collections.
A mundane issue such as space dictates critical decisions in libraries and
archives.
The death knell sounded when archivists began to report that few if
any researchers used constituent mail, so why, they asked, should they
save all of it? The solution was to use random sampling techniques, weed
the series down 80 percent, and call it good. One of the first experiments
in sampling was based on advice and recommendations from experts in
economics, statistics, and mathematics. Further, it was suggested that for
statistical purposes, a scholar would need as little as 2 to 3 percent from
the remaining 20 percent for a valid sample. This procedure involved
assigning random numbers from A Million Random Digits by the Rand
Corporation to each item in a box (a large box filled to capacity holds 2,200
documents). Then documents were removed beginning with a randomly

selected number (McKay 1978). Suffice it to say that by the time the
documents were numbered and removed, the entire box could have been
processed. Other variations were proposed over the years. How many
archivists actually applied these scientific techniques is unknown. There
are reports of archivists consciously selecting “fat files” based on certain
dimensions or retaining every tenth folder, but those approaches would
not be considered true sampling.
Even retaining every tenth item requires a lot of time and discipline.
Invariably, there will be something that catches the eye and falls outside of
the numbering system. Before one knows it, the “keep this” pile outnum-
bers the “toss this” pile. The National Archives and Records Adminis-
tration (NARA) archivists may be the biggest implementers of sampling
and have the most experience. They seem to apply it sparingly, however,
because it is so labor-intensive.
Partly as an experiment, partly in appeasement, the archivists at the

Arizona Historical Foundation responded to Sean Kelly’s repeated pleas
to retain constituent correspondence. This series may also be another
notable exception in the Goldwater papers: it was largely organized by
topic. Many of the congressional officeholders in the generations after
Goldwater organized these files by date or alphabetically by sender. The
archivists who worked on the Goldwater papers did not weed. They saved
it all. They even tried to describe constituent service in ways that might
have meaning for political scientists. Here are excerpts from the two major
subseries.
Projects and Programs generally capture local and regional

grants, awards, proposals; technical and annual reports, court
claims and legal briefs; federal, state, county, and city staff and
agency initiatives; staff special projects; and programs of special
interest to Senator Goldwater. This section is also a rich source
of local history. It is a rich source of gray literature—unpub-
lished, difficult to locate materials such as technical reports, pre-
prints, fact sheets, patents, working papers, meeting minutes, busi-
ness documents, newsletters, bulletins, symposia, standards, white

papers, proceedings and unpublished works. As such, it has high
research value.
Issue Mail is organized by topic and can be considered “current

events and issues at a glance,” i.e., Watergate, Panama Canal,
or Taiwan Treaty. This sub-series is organized by topic primarily
established by Goldwater’s office staff and often noted in the upper
right hand corner of Goldwater’s response letter. These terms were
incorporated in folder titles wherever possible. Like-topics were
combined within congressional sessions when feasible. Bulk deter-
mined what could be consolidated and what could stand alone.
Large runs of correspondence were left under their original desig-
nations. Note that these descriptive titles shifted over time and are
not uniform, i.e. Immigration, Border Patrol, Illegal Aliens, Farm
Labor, Mexican Nationals. Unless noted, all issue mail has been
retained. Where possible all enclosures have been retained. If a
form letter was attached to orphaned documents, it was considered
Issue Mail. Note that form letters have been aggregated within the
Series: V. Administrative Series. Although Goldwater’s positions
are represented in form letters, many of the responses found here
deviate from a set script and often reflect a personal quality.
Surprisingly, this series has been used in depth from the moment it was
made available in the Goldwater papers. Subjects retrieved include urban
growth, school funding, energy, senior citizens, pollution/environment,
historic preservation, immigration, equal rights, the Vietnam War, Water-
gate, and water resources. Ironically, the biggest users of the constituent
service series thus far have been historians.
The Case for Case Files

Case files testify to those people who have been lost in the system, fallen
through the cracks, or left to fend for themselves: the poor, the incarcer-
ated, the sick, the illiterate, the mentally ill, the homeless, the veterans,
the elderly, the addicted, the single mothers, the abandoned children, the
disabled, the refugees, the jobless, and the alienated from society. They are
all here and they beg to be studied. This is evidence of what happens when
the system fails, when safety nets disappear, when the funding stops, or
when programs do not work. It also documents the impact of new policies
such as immigration and welfare reform.
Many case files never reach a repository because they are considered
sensitive and are destroyed once the case is closed. The files that do reach
a repository are often destroyed for the same reason. The Goldwater case
files are an exception likely because they were overlooked or were not
recognized as case files when they arrived. These records are voluminous,
largely intact, and date from 1969 to 1986. They contain every kind of
personal information imaginable: social security numbers, health records,
social service files, mental health files, prison records, military records,
financial records, immigration records, and personnel files. They were
discovered adjacent to the rows of constituent mail. A description from
the scope and content note follows.
Numerous case files (60 linear feet) from the 91st–99th Congress
were discovered among this material. Although not listed here,
individual case files have been retained for research. They are orga-
nized according to the federal agency or department to which the
cases were referred. These files are considered sensitive. Please
contact the archivist for permissions and access.
The archivists working on the Goldwater papers took a deep breath and
did not destroy them. Here is why.
If “sensitive” (defined here as personal information) were an objective

criterion to destroy or redact records, then medical research would come
to a screeching halt. Historically, access has been granted to sensitive
materials under certain conditions for the purposes of scholarly research.
On that basis and within those access parameters, these case files were
retained. Researchers sign a special permissions form requiring them to
observe confidentiality and ensure that no personal data will be reported
or identified in any publication. For other exceptions to the rule, Senator
Edmund Muskie’s case files have been retained at Bates College and
Senator Masayuki Matsunaga’s case files are included in his collection at

the University of Hawaii. Matsunaga’s case files are described as follows:
Case Files—4 record center boxes, 2 document cases
This series is restricted for use except by permission of

the Archives staff. The files include cases handled by the
Matsunaga office relating to constituents’ personal problems
chiefly concerning employment, military service, and veterans’
and Social Security benefits.
For all the members of the American Political Science Association

(APSA) New Political Science section, hear our plea. This material
is ready-made for you. Congressional case files will fast-forward your
mission statement goal of making “the study of politics relevant to the
struggle for a better world.” See the list of Goldwater case files in the
appendices.
The Container List

The final section of the finding aid is the container list, which represents
the bulk of the document. This section lists the contents of a collection
at the box and folder levels. Note that the finding aids give no indication
of the size of each folder. It could contain one to more than one hundred
documents. The folder title represents the subject matter of the documents
within. Dates on the folder reflect the date range of those documents. Items
without dates are usually listed as “undated” or no date (n.d.)
Descriptive detail within a container list varies from institution to insti-

tution. This reflects internal policies and guidelines and represents a
fine balance between what researchers need and want (more description)
compared to the resources that are available (never enough). The more
description, the easier and faster the navigation. Time is the rate-limiting
factor. The archivist’s goal is to get the researcher “close enough” without
searching through dozens of boxes. This increases researcher efficiency
while minimizing wear and tear on the documents.
If a finding aid is in Portable Document Format (PDF) format, use the

CTRL-F or Command-F function and enter keywords to locate a person,
place, topic, date, or type of material, such as speeches. If the finding aid
is in Encoded Archival Description (EAD), the entire document can be
searched from the first-page menu. Table 3 is a sample of the container
list from the 1964 presidential campaign series in the Goldwater papers.
Even this small slice has many things to count.
Table 3. Container list from the 1964 presidential campaign series in the Gold-
water papers.
Search Strategies
If collections are processed, catalogued, and online, Google will likely
capture them, but they may not be obvious. Some links will direct the
searcher to a PDF, others to a web page, and still others to a database. More
university and research library catalogs now incorporate direct searches
for archival material. University and college libraries also have special
collections departments with their own web pages that may list holdings
separate from the library catalog. Most nonacademic repositories have
robust websites that allow for searching collections.
Since the mid-1990s, great strides have been made by the archival
community to provide access to finding aids online. These can be
displayed as the ubiquitous PDF files or converted into EAD. The latter
is considered the gold standard for discovering archival collections on the
Internet. This special tagging software allows for searching all or sections
of the finding aid. Until recently, only the larger institutions or a consor-
tium of repositories could leverage the funding to support specially trained
staff to convert Word documents into EAD and the technical infrastructure
that is required to upload, display, and maintain Internet content.
The appendices will list many archival sources, but currently there is no
reliable “ready reference” for locating specific collections or unprocessed
collections or for finding out the status of someone’s papers (will they be
available, and if so, when and where?). What follows pales in comparison
to the multiple databases political scientists use for data mining and large-
scale computations. Archivists’ needs are relatively simple. They are the
consumers of their own services. They provide references for colleagues,
who provide references in return. When archivists need something, they
usually need it quickly, so they make the most with what is at hand.
Though librarians may wince when they hear this, we go to Google first.
With more and more finding aids coming online, there is a good chance of
capturing what one wants. The advanced search option connects to several
specialized areas. One of these is the U.S. government. This is the best
portal to all things federal. Because congressmen are public figures, the
odds favor hits that will lead somewhere—even if the individual is retired
or deceased. We give Google maybe five minutes. If nothing turns up, we
cut our losses and move right on to WorldCat.
WorldCat is touted as the “world’s largest library catalog,” with access

to 1.4 billion items in ten thousand libraries around the world. This is
humanities heaven. The version we use is the database designed for acad-
emic libraries, not the online subscription for the general public. Every
major library in the United States subscribes to WorldCat. We use it to
search for newspapers, out-of-print and rare books, and archival collec-
tions. We use it to verify citations, review sections of the catalog record

to see how the experts entered data, and see how our manuscripts are
retrieved in a search. The latter is no small thing because of the strict
requirements for original cataloguing—a universe unto itself. WorldCat
has added two intriguing indexes, cleverly called PapersFirst and Proceed-
ingsFirst. They search papers (in all disciplines) from conferences, sympo-
siums, and expositions worldwide—more gray literature that can no
longer hide. On that topic, because it holds a certain fascination, the data-
bases listed on the APSA site for proceedings, presentations, unpublished
material, and works in progress would appear to leave no stone unturned.
Political Research Online is very impressive and illuminating for the ratio
of quantitative to qualitative studies in the discipline. No worries; archives
can easily accommodate both.
The Social Security Death Index contains more than 80 million records.
It is a good source for verifying birth and death dates and where a
person died; depending on the record, a Google map may come up
showing where a person is buried. Would a political scientist need this?
Yes, maybe, if he or she were searching for people to interview or to
verify if someone of interest was deceased. We have used it as a kind
of echo-location for finding descendents and heirs, tracing the where-
abouts of family members connected to collections, addressing copyright
issues, and helping researchers locate these individuals. If we know where
someone died or is buried, we then go to the local telephone directories to
match names and ages. Then we start calling, often with success.
The Society of American Archivists (SAA) staff field a wide variety of

inquiries and make numerous referrals to experts in the field. They also
work with archivists nationwide from every major repository, which gives
them a unique perspective of the profession. SAA has thirteen sections
representing the larger types of archives and major archival topics. There
are thirty roundtables, which are smaller, specialized interest groups. Their
members are well connected and know how to find most things in their
specialties. Each has a chair and steering committee that can speak for the
group. Each section and roundtable has a listserv, and one does not need
to be a member to join. The Congressional Papers Roundtable would be
the perfect place to post a query about the status of a certain congressional
collection.
A�Parting�Gift
If we have not yet made our case that congressional collections are a supe-
rior match for the research needs of political scientists, then let us conclude
with a list of what one would be missing by staying away from primary
sources. This also applies to certain organized sections of APSA. Based
on a quick review of the mission descriptions, the members of the politics
and religion, legislative studies, political organizations and parties, qual-
itative and multimethod research, political communication, politics and
history, political science education, psychology and politics, new political
science, and public policy sections would appear to be candidates inclined
to use archives. Ten possibilities out of forty-one sections is better than
might be expected.
What follows are descriptions of types of archival collections beyond
the congressional. They deserve undivided attention. Although they are
not generally thought of as political papers, these collections are unexpect-
edly rich in political content. Usually they are fully processed and cata-
logued, and the finding aids can be found online. Compared to the congres-
sional collections, these collections are well organized, fairly predictable,
and therefore easier to use. They have one thing in common: data.
Personal papers (general manuscript collections). Think of names,

people behind the scenes, peripheral players that in aggregate
may reveal important patterns and insights. Campaign managers,
state party chairmen, convention delegates, community orga-
nizers, entrepreneurs, philanthropists, business leaders, grassroots
activists, writers, journalists (especially those who covered the
politics/government beat), party organizers, lobbyists, ambas-
sadors, losing candidates, retired faculty, lawyers, judges, promi-

nent political scientists.
Local and state government. Governors’ and state legislators’

papers; state agencies; judicial papers; attorney general records;
cities and towns submitting data for grants, funding infrastructure,
or fighting regulations. SAA has a government records section.
Organizations. Think professional, nonprofit, and interest groups.

Chambers of Commerce, banks, industry and professional asso-
ciations, recall and referendum groups, universities, home-
owners’ associations, causes/advocacy/ethnic groups, neighbor-
hood watches, political parties, environmental groups, unions.
SAA has a labor archives roundtable and a college and university
archives section.
Church archives. Think sermons, newsletters, press releases,

rallies, petitions, radio and television programs, social-commu-
nity-political initiatives. Requests for how to find these and how
to get in were frequent enough that we generated a list of church
archives in the Phoenix metropolitan area. With prior arrange-
ments, they generally allow researchers in. Their national head-
quarters often will do the same. The SAA has an archivists of reli-
gious collections section.
Businesses. Think influential, locally owned, multigenerational,

ethnic or local icons (cannot do business without politics). Public
relations and marketing firms, pollsters, car dealerships, trucking,
grocery stores, health care industry, restaurants, construction,
lobby firms, paving, beverage distributors, refineries, mines, utility
companies, importers/exporters, newspaper publishers. SAA has a
business archives section and a science, technology, and health care
roundtable.
Military. Think retired brass, all service branches, plus academy

records. West Point, Annapolis, and the Air Force Academy are
well connected to the Armed Services Committee, and they know
how to work members on the Hill.
This is but a taste. The possibilities are endless and are limited only by
one’s imagination. We will spread the word about how much fun political
scientists are to work with. (Endorsements do not get better that that!) So,
what is standing in the way? Get out there and plan a visit to the nearest
archives.
Endnotes
1. This quote speaks volumes because it comes from faculty, not undergrad-
uates. It clearly illustrates the gulf between the minority of political scien-
tists who use archival materials and the majority of those who do not.
When a discipline limits its research to one or two genres (i.e., govern-
ment documents or newspapers), it lends credibility to the Kelly-Frisch
arguments that the data have become stale.
2. Backlogs are a blight. Greene and Meissner’s landmark paper in 2005
was a call to arms for archivists to rethink the way they processed large
contemporary collections because backlogs were “hurting the profes-
sion.” Their statistics demonstrate the depth and breadth of a problem
that was sixty years in the making. In one survey, more than 51 percent
of repositories reported that the collections that were processed in the
previous year were done in direct response to donor dissatisfaction. Their
surgical dissection of the issues and the steps to confront them were
nothing short of revolutionary. The shift from “serving the needs of collec-
tions” to serving the needs of users was dramatic. Federal grants became
available to institutions that were willing to innovate and apply MPLP
to a wide variety of collections. Then Greene took things a step further
and provided a model for deaccessioning unprocessed collections on a
grand scale. If the collection did not fit the mission or was outside of
the collecting scope, he transferred it to a repository that could use and
support it. This was a chance to reunite split collections, swap unprocessed
collections for collections that were a better fit, and look for ways that
decreased competition and got materials into the hands of users more
quickly. Almost overnight, repositories began listing unprocessed collec-
tions and making them available to researchers. Deaccessioning, though
it is not done casually, is now commonplace. Certain federal processing
grants require proof that deaccessioning is a part of a repository’s manage-
ment plan. Transparency is the order of the day. A number of assumptions
have been tested and have fallen by the wayside. There is no turning back.
3. Despite the archival definition, the notion of research value tied to use
is not fully embraced within the archives profession. This contributes
to a dearth of archival metrics in the literature. Many repositories count
patrons, items pulled, and copies made, but few seem to record what
manuscripts were used. Unofficially, it would appear that Mo Udall’s
collection is requested on average twelve times a year, Carl Hayden’s

and John Rhodes’s are pulled five to seven times annually, Dennis
DeConcini’s and Paul Fannin’s are requested on average four times a
year, and Barry Goldwater’s papers are pulled eighteen to twenty times
a year. Depending on the repository, any of these numbers could demon-
strate high use. I posted this query to the Archivists listserv to see if
use had been codified since Aronsson’s article: “If (actual and poten-
tial) use is a measure of research value, how would you measure low,
moderate or high use for your processed manuscripts?” The following
responses are revealing: “Our librarian colleagues deem use (check-out)
to be an important measure of a book’s worth; no checkouts means it is
a book worthy of discarding as no one is using it. In archives, however,
this is a major error…I think your premise is wrong that we can assign
low, moderate, high values to archival collection.” And this: “Use of our
legislative collections is low and has stayed within a fairly stable range
for some years. Though we hope use will increase substantially with our
upcoming move to a new facility in the heart of our campus, we will not
be affected if it does not. Many of our collections are currently closed
to research and some that I collected over ten years ago will not open
until long after I’ve retired. We are building a base of research collec-
tions for future generations of scholars. The need for metrics has never
been greater. There are interest groups researching and actively imple-
menting this. Archives and special collections are among the most misun-
derstood units in library systems—closed, mysterious…special. The stuff
defies routine cataloging, reading rooms are not highly trafficked (by
design), and some items require the use of gloves (oh my). Private and
rarified compared to the pedestrian stacks. Metrics are the lingua franca
of libraries. When you cannot translate your work or the worth of your
collections into numbers, you lose.”
4. In 2003, Ellen Proxmire held a press conference because she wanted to
keep her husband’s legacy alive after his years of seclusion. (He was
in a nursing home and suffering from Alzheimer’s disease.) She wanted
people to know that he was still alive. Senator William Proxmire had
donated his papers (360 linear feet) thirteen years earlier, but they had
remained unprocessed at the Wisconsin Historical Society. During the
press conference, Proxmire described it as a “big, big problem” because
the papers were “just sitting there in boxes” unused. Their creator had
become increasingly unknown to the very constituency he had served.
Gaylord Nelson, a former Wisconsin governor and later a U.S. senator,
had given the society seven hundred banker boxes of his papers years
earlier. Like Proxmire’s papers, they remained unprocessed. In an inter-
view, the director of the Wisconsin Historical Society stated that “this was
no crisis” and that these situations were not unusual in archives, espe-
cially during budget cuts. Since then, it appears that both collections have
been processed. The scope and content note in Proxmire’s finding aid is
a cautionary tale and illustrates every conceivable problem with office
records management, serial deposits, restrictions, and donor expectations.
It is rare for an archivist to describe a collection as “disappointing” in a
scope and content note, but in the finding aid for Proxmire’s papers, this
is clearly stated and the reasons why are listed. Proxmire and Goldwater
were contemporaries and frequent sparring partners—both quick-witted
and fearless. In many ways, they were well matched save for their polit-
ical ideologies. Scholars have been denied the chance to fully analyze,
contrast, and compare the two. This is a real loss, and scholars are all the
poorer for it.
5. The Goldwater collection began with 1,200 boxes (not counting artifacts
and memorabilia). Here are the numbers (bulk decreased by 30 percent)
after the processing was done:
1180 linear feet; 970 boxes; 1.14 million documents

8,000 unmounted photographs; 1,500 negatives; 5,000 slides; 110
photo albums
107 news clipping scrapbooks; 480 reels of microfilm
1,028 film reels, cassettes, and tapes
896 pages (finding aid)
125+ years (1880s–2008) of Arizona and U.S. history
6. More than half of the reference requests for Goldwater’s papers come
from out-of-state researchers. Typically, they are PhD candidates, faculty
seeking tenure, or postdoctorates and fellows. Planning the first visit aver-
ages four e-mail exchanges before settling on the specific boxes/folders
to pull and identifying relevant materials in other collections. Sometimes
e-mail fails altogether and one party calls the other. This is not time
wasted. Approximately thirty-five to forty boxes are pulled for the first
visit, which lasts three to five days. These scholars tend to spend eight-
plus hours each day in the reading room. On average, historians leave
our repository with more than five hundred copies and political scientists
fewer than three hundred copies. What follows is a typical e-mail query
and response. This is between a well-traveled David Parker asking about
the Harrison Schmitt papers at the University of New Mexico (UNM) and
an enthusiastic Rose Diaz, the director of the UNM Political Archives.
Note the level of detail in both communications. Used with their permis-
sion.
Dear Ms. Diaz:

At the moment, I’m working on a book project which focuses on
Senate incumbents who lose re-election, and I’m interested in exam-
ining Harrison Schmitt’s 1982 campaign against Jeff Bingaman.
Before visiting a collection, I like to get a sense for the scope of mate-
rials available to researchers.
Given my topic, I’m particularly interested in items related to the
1982 re-election campaign. These would include correspondence
concerning the campaign, press releases from the campaign, strategy
memos, polls, campaign advertisements (video, radio, newspaper,
etc), campaign briefing books, campaign brochures, and the like.
I’m also interested in items related to the Senator’s travel back to New
Mexico and how he communicated with constituents. Copies of the
Senator’s travel schedules, for example, would be ideal. I’d also be
interested in looking at some items from the 1976 race, in particular
polling data for comparison with polls from 1982.
Would you be able to give me some sense of whether the collections
contain items of this nature, and approximately how many boxes of
the collection might be related to the items of interest?
Thanks very much for your time and assistance in this matter.
I look forward to the possibility of a visit in the near future.
David C. W. Parker,
Assistant Professor,
Department of Political Science
RESPONSE:
Dear Dr. Parker:

I was happy to receive your message regarding research in the
Harrison Schmitt papers. At present this collection has only been
preliminarily inventoried and is not generally open to the public.
However, since some of the materials you would be interested in
seeing can be readily identified in the preliminary inventory, I have
received permission from our Director to allow you access to such
materials during your visit to NM.
At present, Charlotte Walters (a staff member) has worked with the
collection and is best equipped to address the availability of mate-
rials contained within your reference questions. I will forward your
request to her and she will be in touch with you very quickly. She
is also able to provide you with electronic copies of the preliminary
inventory from which to select research materials for viewing. While
we are allowing access to the unprocessed collection, we are not able
to provide extensive reference services prior to your visit since, most
often, such preliminary inventories cover very large sections of the
collection. We can provide you only the limited information currently
available to us.
We are thrilled that this collection is on your list to research and look
forward to meeting you in the near future. Also, we are not located on
campus as the Congressional collections are far too big to work with in
library prime real estate, so you will have to come to the processing site
located just a mile from campus (really easy parking, too!). Charlotte
can fill you in on local hotels and our location when you have a better
idea about your trip.
Sincerely,
Rose Diaz, PhD,
Research Historian,
UNM/CSWR/Political Archives
7. The experience with Senator DeConcini and his deed of gift left a lasting
impression. His restrictions perhaps had outlived their usefulness because
Frisch and Kelly were precisely the kind of scholars he hoped would use
his papers. Since then, we have approached other donors with onerous
deed-of-gift restrictions unrelated to federal code. We recently made the
case for a collection with high research value (political but not congres-
sional) that had significantly deteriorated over time. We were not only
successful in having the fifty-year restrictions lifted but also persuaded
the donor to allow us to transfer it to a repository that was more geograph-
ically appropriate.
Bibliography
Abraham, Terry, Stephen E. Balzarini, and Anne Frantilla. 1985. “What Is

Backlog Is Prologue: A Measurement of Archival Processing.” The Amer-
ican Archivist 48(2): 31–42.
American Library Association & Society of American Archivists. 2009. The
ALA/SAA Joint Statement on Access to Research Materials in Archives and
Special Collections Libraries. http://www.archivists.org/statements/ala-
saa-access09.asp.
Aronsson, Patricia. 1984. “Appraisal of Twentieth Century Congressional
Collections.” In Archival Choices, edited by N. Pease, 81–104. Lexington,
MA: Lexington Books.
Barzun, Jacques, and Henry F. Graff, 2004. The Modern Researcher. 6th ed.
Belmont, CA: Wadsworth/Thomson Press.
Blouin, Francis X. Jr. 1999. “Archivists, Mediation, and Constructs of Social
Memory.” Archival Issues 24(2): 101–112.
Brady, Rep. 2008. “Expressing Sense of Congress That Members’ Congres-
sional Papers Should Be Properly Maintained.” Congressional Record 154
(March 5): H1254.
Brooks, Philip C. 1951. “Archivists and Their Colleagues: Common Denom-
inators.” American Archivist 14(1): 33–45.
Case, Donald O. 2002. Looking for Information: A Survey of Research on
Information Seeking, Needs, and Behavior. San Diego, CA: Academic
Press.
Circular 92: Copyright Law of the United States and Related Laws Contained
in Title 17 of the United States Code. 2007. Washington, DC: Government
Printing Office.
Cook, Terry. 2001. “Archival Science and Postmodernism: New Formulations
for Old Concepts. Archival Science 1: 3–24.
———. 2009. “The Archive(s) Is a Foreign Country: Historians, Archivists,
and the Changing Archival Landscape.” The Canadian Historical Review
90(3): 497–534.
DeConcini, Dennis. Papers, 1944–2003. University of Arizona

Library Special Collections. http://www.library.arizona.edu/speccoll/find-
ingaid/deconcini_papers/.
“Encoded Archival Description Implementors.” 2009. Society of American
Archivists. http://www.archivists.org/saagroups/ead/implementors.html.
Freeman, Elise T. 1984. “In the Eye of the Beholder: Archives Administration
from the User’s Point of View.” The American Archivist 47(2): 111–123.
Frisch, Scott A., and Sean Q Kelly. 2003. “Don’t Have the Data? Just Make
Them Up! Congressional Archives as Untapped Data Sources.” P.S. Polit-
ical Science and Politics 36(2): 221–224.
———. 2004. “Political Scientists: Strangers in a Wonderland.” Congres-
sional Papers Roundtable Newsletter (July): 5–7.
———. 2005. “Five Reasons to Consider Taking the Road Less Traveled.”
LSS Newsletter, Extension of Remarks: 28–33. http://www. apsanet.org/
~lss/Newsletter/july05/Frisch05.pdf.
———. 2009. “What Archivists Need to Know About Political Scientists.”
In An American Political Archives Reader, edited by G. R. Gray, R. J.
Melvin, and K. D. Paul, 401–418. Lanham, MD: Scarecrow Press.
Greene, Mark A. 1994. “Appraisal of Congressional Records at the Minnesota
Historical Society: A Case Study.” Archival Issues 19: 31–43.
———. 2002. “The Power of Meaning: The Archival Mission in the Post-
modern Age.” The American Archivist 65(1): 42–55.
———. 2003–2004. “The Messy Business of Remembering: History,
Memory and Archives.” Archival Issues 28(2): 95–103.
———. 2006. “I’ve Deaccessioned and Lived to Tell About It: Confessions
of an Unrepentant Reappraiser.” Archival Issues 30(1): 17–18.
Greene, Mark A., and Dennis Meissner. 2005. “More Product, Less Process:
Revamping Traditional Archival Processing.” The American Archivist
68(2): 208–263.
Haller, Uli. 1987. “Variations in the Processing Rates on the Magnuson and
Jackson Senatorial Papers.” The American Archivist 50(1): 100–109.
Ham, Gerald. 1984. “Archival Choices: Managing the Historical Record in an
Age of Abundance.” The American Archivist 47: 11–22.
Harris, Douglas B. 2009. “Recovering History and Discovering Data in the

Archives: An Alternative Mode of Research for Congress Scholars.” In An
American Political Archives Reader, edited by G. R. Gray, R. J. Melvin,
and K. D. Paul, 429–440. Lanham, MD: Scarecrow Press.
Hartnett, Cassandra. 2009. “Together We Can Save Congress.” Presen-
tation on Government Documents at the August 2009 CPR Program
GODORT Chair, Gov Docs and Women’s Studies Librarian, Univer-
sity of Washington. http://www.archivists.org/saagroups/cpr/publica-
tions/hartnett-austin09.pdf.
Hirtle, Peter B., Emily Hudson, and Andrew T. Kenyon. 2009. Copyright
and Cultural Institutions: Guidelines for Digitization for U.S. Libraries,
Archives, and Museums. Ithaca, NY: Cornell University Library.
Kaplan, Alice Y. 1990. “Working in the Archives.” Yale French Studies 77:
103–116.
Kaplan, Elisabeth, 2000. “We Are What We Collect, We Collect What We
Are: Archives and the Construction of Identity.” The American Archivist
63(1): 143–151.
Light, Michelle, and Tom Hyry. 2002. “Colophons and Annotations: New
Directions for the Finding Aid.” The American Archivist 65(2): 216–230.
Louisiana State University, Shreveport. 2009. “Libraries and Archives.”
http://www.lsus.edu/library/archives/resources/comparison.htm.
Lucas, Lydia. 1978. “Managing Congressional Papers: A Repository View.”
The American Archivist 41(3): 275–280.
McCloskey, Paul N. Papers Inventory. Hoover Institution Library and
Archives. http://www.oac.cdlib.org/findaid/ark:/13030/kt1s20243r/.
McKay, Eleanor. 1978. “Random Sampling Techniques: A Method of
Reducing Large, Homogeneous Series in Congressional Papers.” The
American Archivist 41(3): 281–289.
Miller, Cynthia Pease. 2008. Managing Congressional Collections. Chicago:
Society of American Archivists.
Miller, Frederic M. 1990. Arranging and Describing Archives and Manu-
scripts. Chicago: The Society of American Archivists Press.
Morris, Thomas Gayle. Register of Congressional Papers. New Mexico State
University Library, Archives and Special Collections. http://rmoa.unm.
edu/docviewer.php?docId = nmlcu1ms31.xml
New Frontiers in Gray Literature. 1999. Fourth International Conference on

Gray Literature, October 4–5. Washington, DC, and Amsterdam: GrayNet,
1999.
New Mexico State University. 2008. “NMSU Announces Senator Pete V.
Domenici Legacy Project.” http://lib.nmsu.edu/aboutlib/libnews/Domeni-
ciCommittees.shtml.
O’Toole, James M., and Richard J. Cox. 2006. Archives Fundamentals Series
II: Understanding Archives and Manuscripts. Chicago: The Society of
American Archivists Press.
Pearce-Moses, Richard. 2009. Glossary of Archival and Records Terminology.
http://www.archivists.org/glossary/list.asp.
Richardson, Jeff. 2009. “BP Donates $1 million to Archive Former Sen. Ted
Stevens’ Papers.” Fairbanks Daily News Miner, November 19, p. A1.
Rudd, Eldon. Papers. Arizona State University Libraries, Archives and
Special Collections. http://www.azarchivesonline.org/view/asulib:14763
8.
Schram, Sanford F., and Brian Caterino. 2006. Making Political Science
Matter: Debating Knowledge, Research, and Method. New York: New
York University Press.
Skemer, Don C. 1991. “Drifting Disciplines, Enduring Record: Political
Science and the Use of Archives.” The American Archivist 54(3): 356–
368.
Society of American Archivists. Describing Archives: A Content Standard.
2004. Chicago: The Society of American Archivists Press.
Straw, Joseph E. 1993. “Convergence or Divergence? A Study of the Infor-
mation Seeking Behavior of Historians and Political Scientists.” Master’s
thesis, Kent State University School of Library and Information Science,
Kent, OH.
Unekis, Joseph K. 1996. “Researching Congress.” P.S., Political Science and
Politics 29(4): 719–722.
Whitaker, Linda A. 2005. “Political Papers, the Archivist and You.”
LSS Newsletter 28(2). http://www.apsanet.org/~lss/Newsletter/july05/
Whitaker05.pdf.
Wikipedia. 2009. “Seniority in the United States Senate.” http://en.wikipedia.
org/wiki/Seniority_in_the_United_States_Senate.
Yakel, Elizabeth, and Deborah A. Torres. 2003. “AI: Archival Intelligence and
User Expertise.” The American Archivist 66(1): 51–78.
Yakel, Elizabeth, and Jeannette Bastian. 2005. “Special Report: Grad-
uate Archival Education and the A*Census.” http://www.archivists.org/a-
census/reports/YakelBastian-ACENSUS.pdf.
Part�II
Political�Science
in�the�Archives
Chapter 4
Why�Archives?
David C. W. Parker
When I decided to go to graduate school, I knew I wanted to study

the representative process—specifically, campaigns. How do members of
Congress navigate their relationships with constituents to win reelection?
How do constituents view their members of Congress and their repre-
sentative activities? I worked on a mayoral, presidential, and two Senate
campaigns in 1996 and saw public servants at multiple levels of govern-
ment make the case that they understood their constituents, had done good
things for them, and deserved to be elected. Campaigns, to my way of
thinking, matter—they are the heart and soul of the U.S. political system.
They legitimate a political system that delegates political power rather
than allowing the political masses to exercise it directly in the policy-
making process.
Imagine my surprise when I found out that much of political science

conventional wisdom dismissed the role of campaigns. Forecasting
models demonstrated that by using approval ratings and economic growth,

election outcomes could be accurately forecast months in advance. Parti-
sanship, indeed, seemed to predict most elections. All of the money,
time, and effort members of Congress put into connecting to voters and
constituents—according to this literature—was a waste of time, energy,
and effort. To quote Shakespeare’s Macbeth, campaigns were full of sound
and fury, signifying nothing.
This did not sit well with me, particularly with the increasing turn to
rational choice models of human behavior in the social sciences. Political
actors are rational actors—they minimize their costs and maximize their
benefits. If anyone could be depicted accurately as rational, officeholders
had to be it. So why would officeholders, who want to maximize their
resources, expend all this time, money, and effort in a fruitless endeavor?
Something did not make sense, and I was bound to figure out what it was.
I believed, very simply, that the actions of officeholders—particularly

members of Congress—mattered to constituents. Members do build rela-
tionships with the people they represent, and those relationships help them
get reelected. Campaigns remind voters about those connections, and I
knew that with sustained effort and attention, the effects of campaigns
on voters could be demonstrated plainly. Ultimately, some incumbents
survive bad economies and partisan wave elections and others do not.
What explains this? I was convinced that the campaign, and more specif-
ically, the activities that members engaged in and the connections they
established with their constituents as communicated during campaigns,
mattered. Campaigns matter because representation matters.
The problem is, how does one observe those relationships with
constituents? Much of the existing political science data did this poorly.
Although the American National Election Studies (ANES) asks questions
concerning media exposure, attention to politics, and the recall of political
advertisements, the studies do not allow for a sustained study of congres-
sional campaigns. The ANES 1988–1992 Pooled Senate Study allows for
the in-depth study of Senate campaigns; unfortunately, the 1988, 1990,
Why Archives? 163
and 1992 election years saw few Senate incumbents actually lose reelec-
tion, and the study is more than twenty years old at this writing. Add to that
the fact that even the ANES Pooled Senate Study—as rich as it is—inter-
viewed respondents after the conclusion of the campaign, not during the
campaign. It is hard to understand the dynamics of an ongoing campaign
based on this design.
My undergraduate training was mostly as a historian. The use of

primary source documents was not a mystery to me. Archival materials,
however, were. My work on the Truman administration’s relations with
Great Britain relied primarily on State Department documents gathered in
the “Foreign Relations of the United States” series. These are bound, typed
documents, all located in one neat, tidy place. My first experience with
archives occurred when I was in graduate school. I was trying to move
conceptually from David Mayhew’s measure of state party institutional-
ization (traditional party organization, or TPO) to a measurement of party
involvement in the campaigns of individual senators in the mid-twen-
tieth century. The idea was simple: if I could show that political parties
in so-called “strong party states” involved themselves more in the daily
operation of Senate campaigns compared to weaker party states, I could
demonstrate that the standard story of party decline and the rise of candi-
date-centered campaigns was more nuanced and complicated. To develop
the correct measure, it seemed sensible to track how parties were involved
in the strategic decision making of Senate campaigns, what their role in
fund-raising was, and whether party issues were central in the campaign
itself. To do this, I turned to archival materials: the papers and oral histo-
ries of U.S. senators who first ran for the Senate during the so-called
decline of party-centered campaigns between 1940 and 1960.
The end result was an incredible experience, both in terms of research

and writing. I received a small grant and traveled to archives in Kentucky,
Idaho, Illinois, and Tennessee. Idaho and Tennessee were considered weak
party states, whereas Kentucky and Illinois were seen as strong party
states. Sure enough, the archival record made plain that the involvement
of parties in the day-to-day operation of U.S. Senate campaigns varied

based upon the party’s overall organizational strength in the state. Future
U.S. senators Estes Kefauver and Frank Church essentially ran their own
campaigns, independent of the established party organization and bosses,
during their campaigns in 1948 and 1956, respectively. Senators Everett
Dirksen and John Sherman Cooper, however, had campaigns that were
tightly integrated with statewide party efforts. The evidence in the archives
varied tremendously, too. In Senator Dirksen’s papers, I found campaign
letterhead that listed all of the statewide Republican candidates for office
together and a letter chiding the Dirksen campaign for not clearing its
meetings in Chicago with the local party organization folks. In Senator
Church’s papers, there is a memo advising speakers stumping the state to
avoid national party issues and focus instead on the personal failings and
record of Church’s opponent, incumbent senator Herman Welker.
The archives provided a rich opportunity for me, the political scien-
tist, to develop a new measure of a theoretical concept while providing
a clearer picture of how Senate campaigns operated in the mid-twentieth
century. The process was messy and dirty—much like politics itself. But
the end result was a much richer, more detailed, and more accurate account
of campaigning—one that that was very different from what would have
emerged if I had relied exclusively on the conventional wisdom that the
political science literature had peddled to graduate students and scholars
over the past forty years. The party-centered campaign had not died in the
mid-twentieth century but was thriving in those places where parties had
always been, and continue to be, strong organizationally and culturally.
After my experience working with archival materials, I was, quite

simply, hooked. My current project on senators who lose reelection returns
to the initial question about representation that drove me to graduate
school in the first place: How do representatives build relationships with
constituents? Consider the following narrative. In 1982, freshman senator
Harrison “Jack” Schmitt lost reelection in an expensive campaign. Two
weeks before Election Day, a political writer for the Albuquerque Journal
Why Archives? 165
told a reporter from The New York Times that Schmitt was in trouble
because of disenchantment with Ronald Reagan and economic problems
in the state (Roberts 1982, B12). This, of course, represented the standard
political science line: midterm elections are a referendum on the sitting
president, and congressional incumbents bear the brunt of voter displea-
sure. Quite simply put, Schmitt was in trouble because he was a stalwart
Reaganite in a generally Democratic state and the economy was bad. Of
course he would lose.
But there is an alternate and potentially more compelling narrative to be
teased out of the archival record. Consider first that although New Mexico
was—and still is—a Democratic-leaning state, Republican senator Pete
Domenici successfully represented the state for more than thirty years,
serving from 1973 through 2008. Why did voters not punish Domenici in
1978 when he ran for his first reelection and congressional incumbents
in both parties were thrown out of office? Schmitt’s loss was about more
than Reagan and New Mexico’s economy alone. It is, quite simply, a story
of an incomplete representational connection with his constituents.
Schmitt, on the surface, did all the “right” things one expects of a
freshman senator. He traveled back home frequently. He drove around
the state in a red, beat-up Ford pickup truck. He met with constituents
around the state. He had a compelling biography: a bright kid from a small
town earned a Harvard PhD and was selected for NASA’s Apollo program
because of his geological expertise. Schmitt was the last astronaut to walk
on the moon. Quite simply put, he was a hero and an icon in New Mexico.
He should have won.
But what he did not do—and what Domenici did well—was forge a
strong, positive connection with those constituents. In September of 1982,
two months before Schmitt lost, the evidence of this weak constituent rela-
tionship abounds in survey data commissioned by the Schmitt campaign.
When they were asked why they were not voting for Schmitt, only 13
percent of the respondents volunteered that it was because Schmitt was
a Republican. Another 7 percent mentioned the economy, and 2 percent
mentioned Schmitt’s conservative ideology. Eleven percent indicated that

Schmitt had done a poor job and had not represented the state well, another
11 percent indicated that he was ineffective, and another 11 percent noted
that he was not terribly concerned about people.1 Nearly 45 percent of
the responses had little to do with the macro-political environment that
political scientists say explains voting behavior in congressional races.
In a campaign that was decided by less than eight percentage points, it
is entirely plausible to suggest that Senator Schmitt lost not because of
Ronald Reagan and the economy but because he was not seen as an effec-
tive, dedicated, and compassionate representative of New Mexicans. And
Schmitt’s papers—located at the Center for Southwest Research on the
campus of the University of New Mexico—helped greatly in telling the
richer, more nuanced, and better story.
In the chapters that follow, scholars from a variety of subfields of Amer-

ican politics illustrate their experiences working with archival documents.
They demonstrate from their personal experience how archival research
has improved the quality of their research and improved the depth of their
understanding of a variety of research questions. The chapters also provide
insight into the different kinds of political papers collections that are avail-
able to researchers and the considerations that are unique to each kind of
collection. Feel free to move directly to a chapter that addresses a partic-
ular kind of collection, but take some time to browse the other chapters for
clues about how other kinds of collections might improve research. One of
the keys of good archival research is to remain aware of the serendipitous
opportunities that may present themselves along the road.
Why Archives? 167
Endnotes
1. Tarrance and Associates, September 1982, “A Survey of Voter Attitudes

in the State of New Mexico,” box 604, Harrison “Jack” Schmitt Papers,
Center for Southwest Research, University of New Mexico.
References
Mayhew, David R. 1986. Placing Parties in American Politics. Princeton, NJ:

Princeton University Press.
Roberts, Steven V. 1982. “Schmitt, Astronaut Now in Senate, Struggles to
Hold New Mexico Seat.” The New York Times, October 21, B12.
Chapter 5
If�This�Is�Tuesday,
It�Must�Be�Albuquerque
Using�Archives�to�Research
Congressional�Campaigns
David C. W. Parker
Many of the chapters in this volume address how congressional archives

can help illuminate the institution of Congress: how party leadership
operates, which members receive their desired committee assignments,
and the process of presidential and congressional bargaining.1 By and
large, each chapter illustrates an important methodological point: impre-
cise or improper measurement of concepts can influence one’s substan-
tive findings. The use of archives yields new measurements of theoretical
concepts, and oftentimes these new measures provide better conceptual
fits than do more readily accessible data such as roll call votes, campaign
finance dollars, or election results.
My research uses congressional archives not only to provide new

measures of well-established theoretical concepts but to lift the veil off
of what two scholars call the black box of congressional politics: the
campaign (Coleman and Manna 2000). Much of what is known about
congressional campaigns is observed from the outside—in the brochures
that are produced, the campaign advertisements that are aired, the polls
that are taken by media organizations, and the articles that are written
in newspapers. Less is known, however, about the campaign inner-work-
ings that allocate particular resources and the ways in which these deci-
sions may (or may not) influence election outcomes. More broadly, if
one is to better explain the process of representation, it is absolutely crit-
ical to understand how members of Congress communicate with their
constituents and how they explain their Washington activities to the people
back home. This is a task well-suited to the materials that are available in
congressional archives.
I have used the archival collections of senatorial papers and their elec-
tion challengers at fifteen different locations, from Athens, Georgia, to
Fairbanks, Alaska—as well as many places in between. My research
has focused on Senate campaigns in particular and has addressed two
questions: What is the relationship between political parties and congres-
sional campaigns, and why do incumbent senators lose reelection? In this
chapter, I demonstrate how work in congressional archives can comple-
ment and perhaps even improve upon the method of participant observa-
tion that is strongly advocated by Richard Fenno. Second, I show how
the materials that are commonly available in the congressional collections
yield new empirical measures for political scientists, which aid in both the
process of testing old theories and the construction of new ones. Finally, I
provide tips for avoiding some of the common pitfalls the aspiring archival
researcher faces when choosing to engage in this intense, yet rewarding,
method of inquiry.
If This Is Tuesday, It Must Be Albuquerque 171
The�Method�of�Participant�Observation�Versus
Post–Participant�Observation
Richard Fenno (1978) revolutionized the study of Congress when he
charged political scientists to get out of Washington and into the district to
better understand the process of representation. Although Fenno’s work
has been widely cited and appreciated by the discipline, few political
scientists have taken up Fenno’s gauntlet more than two decades after
the publication of Home Style. As Fenno plainly admitted, the work of
participant observation is mentally exhausting and difficult to accom-
plish in the academic world of “publish or perish” (Fenno 1978, 219–
296; Fenno 2007, 76–78). It is also hard to fund.2 Nevertheless, Fenno’s
work has provided congressional scholars with innumerable insights into
how members think about the puzzle of representation. Most notably,
Fenno (1978) offered the concentric circles of constituency and explicated
the representational styles members communicate to their constituents:
“one of us,” policy expert, and constituent servant. Subsequent empirical
work confirms the use of these different types of representational styles
in how members communicate with their constituents in press releases,
in speeches on the Senate floor, and in the allocation of official resources
such as the frank, travel, and office expenditures (Parker and Goodman
2009; Hill and Hurley 2002; Goodman and Parker 2010; Yiannakis 1982;
Parker and Goodman n.d.). Unfortunately, few scholars have attempted
the in-depth case study approach that was employed by Fenno—espe-
cially on such a grand scale (but see Frisch and Kelly 2006, 2008).3 Those
scholars who are the most disposed energetically to undertake such work
—the young, untenured assistant professors—are also the ones who have
the least incentive to invest so much time and effort in thick, descriptive
analysis. By the time their work in the field was completed, they would
find themselves denied tenure for the sheer lack of publications!
Although Fenno’s work has uncovered a number of important theoret-

ical and empirical insights, the approach clearly has limitations.4 Fenno’s
visits were episodic and provided him with only a small slice of the district
representational experience. And, although he was a keen observer of poli-

tics and human nature, Fenno could not witness many of the important
events during district travels on the campaign trail. As much as Fenno
strove to blend in as a part of the team, he was still an outsider, prone to
the possibility of self-censoring by politicians and their staffs. Participant
observers will not see certain meetings, and they are not privy to all of
the information that campaigns have when they are making strategic deci-
sions about resources and campaign efforts.
Studying the campaign after the fact—sometimes long after the fact—
provides the researcher with certain advantages over the method of direct
participant observation. First, though censorship may still be a problem
(the archival record may have been sanitized in the process of handing
over records to archivists by staffers eager to protect their boss from the
scrutiny of history), the researcher plumbing archives can have access to
memoranda of conversations written by campaign staffers or the candi-
date during the actual campaign—about events the participant observer
may not have had the opportunity to witness. These memoranda are often
quite blunt and frank in their assessments. For example, a memo written
by none other than New Mexico senator Pete Domenici to Harrison “Jack”
Schmitt in 1976 in no uncertain terms describes the core of Schmitt’s
inability to connect with voters during his campaign against incumbent
Joseph Montoya:
Specifically, I believe that you could spend a very useful three

hours with the book, “Ingratiation,” (available from the UNM
[University of New Mexico] library) to learn a little about making
people like you when they meet you. Being bright, competent, well-
intentioned, and earnest is simply not enough. You must be human.
You must drink or swear or sweat or make mistakes. Perfection
remains a very difficult thing for most of us to identify with.5
It appears these problems persisted during Schmitt’s career in the Senate,

and his inability to foster a “one-of-us” connection with his constituents
may have been in part responsible for his defeat in 1982 at the hands of
New Mexico attorney general Jeff Bingaman.
Second, studying the campaign after the fact allows the passions of
the moment to cool and gives way to perhaps a more objective assess-
ment of events by those who were involved. Although one must recognize
that memories erode with the passage of time, candidates and politicians
who are retired speak more frankly, openly, and willingly about campaign
events that are long past. The archival record provides the materials to
well-inform these interviews with key campaign participants.
Third, the method of participant observation often requires focusing on

a few critical actors. In Fenno’s case, he was interested particularly in
the candidate and his daily activities. But of equal importance in under-
standing the campaign are the staffers running the day-to-day operation
and making decisions on the behalf of the candidate. These decisions can
have important repercussions for the campaign’s outcome and are more
difficult to observe. Fenno, in fact, augmented his participant observa-
tions with subsequent interviews. But, in examining campaign strategy
memos, campaign advertising scripts (including the many drafts), and
candidate correspondence, one can uncover political players who may
never have visited campaign headquarters or the campaign trail.6 Yet,
these are the very people who may have had a substantial influence on
the actual conduct of the campaign and its strategic focus. This often
includes the candidate’s campaign consultants and pollsters, as well as
trusted Washington aides offering advice from afar. The manner in which
a member of Congress campaigns and presents him- or herself at home
is, in part, a reflection of this advice from the very intimates lying at the
core of Fenno’s concentric circles of constituencies. These may also be
the same people who are the hardest to observe by dipping in and out of
the campaign.
None of this is meant to suggest that the method of participant obser-

vation is not useful, but merely to point out the differences between it
and the method of post–participant observation. I use the term post–
participant observation specifically because observing events recorded in

archives precludes direct participation, whereas the participant observer
may watch an event contemporaneous with its occurrence and chose
whether to engage. The main advantage of post–participant observation
is a rare glimpse of the second face of power that researchers frequently
discuss (Bachrach and Baratz 1962). The post–participant observer can
sometimes see the path that was not taken by campaigns by paying close
attention to the archival record: for example, the negative campaign ad
that was drafted, scripted, and shot by the consultant but not aired because
of objections by the candidate; the candidate explicitly asking national
speakers to avoid particular issues that might cause the candidate prob-
lems (as Frank Church did in a memo he wrote during the 1956 campaign
against incumbent Idaho senator Herman Welker); or what the campaign
knew but would not acknowledge publicly (that New Mexico senator
Harrison “Jack” Schmitt had very few credit-claiming opportunities to
tout after three years in office).7 Post–participant observation allows the
researcher to look into the hidden and forgotten corners of the campaign
that are not plainly visible to the participant who is observing for a day
or a long weekend. And this is what makes archival work so rewarding
and fun for the researcher.
New�Answers�to�Old�Questions�And�New�Questions
The method of archival research provides insights into campaigns that the
method of participant observation may not. Just as important, however,
is the ability of the researcher engaged in post–participant observation to
engage in theory building, theory testing, and the refinement of concep-
tual measures. Archival research generates new data with which to test
theories, and through the better measurement of critical concepts, it
can demonstrate why some widely used measures may yield erroneous
conclusions about campaigning and the behavior of politicians. After
spending nearly seven years engaged in archival research on congressional
campaigns, I have reaffirmed some of political science’s conventional
wisdom with new data. I have also made some new discoveries which
question some of that wisdom at the very least. At best, these discoveries
may suggest new avenues of research. I have also used the insights gleaned
from archival work to build a new theory of congressional campaigns. In
no particular order, here are some of the things I have learned from digging
around in congressional archives.
Ideology�Measures�Underestimate�the�Importance�of
Issue�Salience
The use of DW-NOMINATE as a measure of a member’s ideology has
become ubiquitous in congressional studies (Poole and Rosenthal 1997).
I, too, am guilty of this, but archival work serves as a reminder of the
problems that are associated with such a measure. First and foremost,
DW-NOMINATE scores are calculated for members using roll call votes
over the entirety of their careers, so any given member’s ideology appears
stable longitudinally.8 Second, each roll call vote is weighted equally in
the calculation of the DW-NOMINATE estimate. Though the end result
produces a good measure of ideology that has become widely accepted
by congressional scholars, the calculation of DW-NOMINATE scores
requires computational concessions that are not terribly realistic represen-
tations of the political world, particularly in the realm of congressional
campaigns. Archival research affirms this vividly.
Take the case of Arizona senator Barry Goldwater. Goldwater is often

depicted as an unwavering, true conservative believer by both contem-
poraries and historians (Edwards 1995; Matthews 1997).9 Certainly,
Goldwater’s DW-NOMINATE score of .63 over the course of his career
suggests the strength of his convictions. An examination of documents
retrieved from Goldwater’s Senate papers suggests a more complicated
reality. In 1980, Goldwater faced his toughest reelection fight yet, against
a young and relatively conservative Democrat. On a host of issues, busi-
nessman Bill Schulz mirrored Goldwater, particularly on taxes, spending,
and national defense. The problem Goldwater faced, however, was the
rise of the religious right, a group that did not necessarily see eye to
eye with him on social issues. In particular, Goldwater had not estab-
lished a clear stance against abortion politically. Opposed personally to
the practice, he had not ruled out a woman’s right to choose without
government interference. In fact, Goldwater’s wife, Peggy, was an active
member of Planned Parenthood in Arizona, and the couple hosted a fund-
raiser for the organization on October 4 during the 1980 campaign.10
Despite his apparent pro-choice sympathies, in the closing weeks of the
campaign, Goldwater publicly endorsed the Human Life Amendment to
the Constitution banning abortion. He did so in a letter to a supporter and
publicly in a forum sponsored by KOOL-TV and the Arizona Republic
in Phoenix (“Goldwater Creates” 1980).11 Goldwater’s apparent flip-flop
became fodder for a Schulz television advertisement, where the narrator
notes that Goldwater had written to pro-choice organizations on five occa-
sions in the past year promising to vote against a Human Life Amend-
ment.12 Goldwater’s voting record depicts a true conservative warrior,
but his personal papers suggest someone who understood the importance
of political calculation. This underscores that an issue’s salience matters
during a campaign, something which relying upon roll call votes or DW-
NOMINATE scores clearly underestimates.13
The struggle a candidate faces over a particular issue can be illus-

trated by his or her correspondence or memoranda dictated for the record.
Another useful tool that helps researchers understand the visibility and
salience to voters of particular campaign issues is the constituent corre-
spondence which is the bane of every congressional archivist’s existence.
Constituent mail is prolific and oftentimes repetitive. Despite its repeti-
tiveness and volume, a flood of mail on a particular issue can help polit-
ical scientists understand its salience to voters, as well as the intensity of
public opinion on the matter.
Although Georgia senator Herman Talmadge faced a number of prob-

lems of his own making during his 1980 reelection campaign (especially
a slew of ethics charges which led to a Senate vote admonishing him for
financial improprieties in the management of his Senate office), his vote
in favor of the Panama Canal Treaties might have played a role in his
defeat. At the very least, Talmadge received a lot of constituent corre-
spondence on the matter, including letters, petitions, and postcards. A
tally of this correspondence yielded 1,022 opposed to the treaties, four
neutral, and only one in favor. A letter from Carl E. Anderson from Garden
City, Georgia, to Talmadge dated October 25, 1977, is typical of the anti–
Panama Canal Treaty correspondence that was received by the senator.
Anderson wrote,
I was shocked when I learned that our president wanted to turn over
control of OUR CANAL to a leftist Dictator. The more facts I learn
the madder it makes me. Can President Carter and YOU be so out
of touch with the American people, particularly fellow Georgians
that he has allowed these events to progress to this point?14
Garden City is a suburb of Savannah, Georgia, in Chatham County.

Talmadge won Chatham County in 1974, but it is one of only twenty-
nine counties (out of 159) he lost in 1980 to Republican Mack Mattingly.
Other correspondents writing on the canal issue hailed from rural counties
throughout the state that Talmadge ultimately won, but with much lower
margins than in past elections. Despite Talmadge’s sound work on farm
issues as the chairman of the Senate Agriculture Committee, his ethics
woes and support of President Jimmy Carter on Panama may have sunk
his reelection chances. Constituent correspondence is yet another way to
measure the intensity of a constituency’s feelings about an issue, some-
thing that can also be measured with surveys (for example, by asking what
is the most important problem facing the nation or a state). Unfortunately,
few surveys of state or local races done by national polling organizations
provide such a detailed picture of the electorate—and those surveys that
do are usually the exit polls that are taken on Election Day.
Going�Home�Probably�Affects�Election�Outcomes
Another important activity illuminated by congressional archival collec-
tions is a member’s trips back to the state or the district. Richard Fenno
(1978) demonstrated the importance of a member’s particular represen-
tational styles as communicated to constituents on visits back home.
Travel home itself can become an important campaign issue, too. In 1984,
Mitch McConnell successfully made Senator Walter Huddleston’s atten-
dance record and trips outside of Kentucky an issue (Andersen, Ogden,
and Phillips 1984).15 In 1980, Goldwater’s opponent accused him of
neglecting his constituent service duties, noting that he “is rarely seen in
rural areas” and people “no longer call his office because he’s not avail-
able.”16 Goldwater privately agreed in a memo dictated to his administra-
tive assistant, Judy Eisenhower, where he wrestles with the decision to
seek reelection in 1980: “I guess it’s been six years since I’ve been in
places like Holbrook and Winslow and other smaller places around the
State, but that’s been at my own choice.”17
It would be useful to measure a member’s travel home as a predictor

of reelection success. This is difficult to implement in practice. Members
of Congress do get reimbursed for official travel home from House and
Senate accounts designated for this purpose. Reporting requirements,
however, have varied greatly throughout the modern congressional era.
There is also a great deal of variance in how congressional and sena-
torial offices record travel home on vouchers. Some offices are explicit
as to where a senator or member traveled when he or she went home;
others simply forward bills from travel agents or credit card companies,
listing the dates of travel and the destinations of round-trip airline flights.18
Finally, official travel does not capture all travel home. Some travel is paid
by a member’s campaign account, and if the member holds a committee
chairmanship, the committee might pick up the costs of travel back home.
Relying upon the secretary of the Senate’s or the House clerk’s records to
count trips home provides an imperfect measure of a member’s attempts
to stay in touch with his or her constituents.
Fortunately, congressional archives usually provide much richer and

more detailed resources for the scholar attempting to account for member
travel. Members’ lives are carefully scripted and detailed by their office
assistants and schedulers. Often, the researcher will find a member’s offi-
cial appointments calendar, detailing a member’s meetings while he or
she is in Washington, and at minimum, indicating when the member is
traveling back home. Better yet, most offices provide detailed agendas
for their bosses when they are traveling back in the state or district.
These agendas list when flights arrive and depart, what communities the
member will visit during the day, and in which activities he or she will
partake, and they often list key city and county political individuals the
member might expect to meet and see. Ohio senator John Glenn’s office
was meticulous in this regard; not only does the researcher know exactly
where Senator Glenn was at any given moment on any given day, he
or she will also likely have extensive background memos on each event
and remarks prepared for each occasion appended to the senator’s daily
schedule. In fact, Glenn’s staff provided an annual summary listing not
only the total number of appearances made by the senator to Ohio but
also the number of appearances he made in each county throughout the
year. Table 4 details Glenn’s travels between 1987 and 1992, the year he
launched his last reelection campaign. It tells three important stories. First,
it provides evidence confirming the Senate campaign-govern-campaign
cycle noted by Fenno. Glenn made only twenty-two appearances in Ohio
in 1987, immediately following his reelection, with travel home essen-
tially increasing each year until 1992, when Glenn made more than 132
appearances in Ohio—a six-fold increase over 1987. Second, the table also
tells where Glenn traveled in Ohio—and it appears that he did a good job
of getting around the state during his six-year term. Third, the mere exis-
tence of the table in Glenn’s congressional papers tells something about
Glenn as an office manager. He demanded precision, a characteristic that
was undoubtedly honed during his prepolitical career in the military and
the National Aeronautics and Space Administration (NASA).
Table 4. Senator John Glenn and going home, 1987–1992.
Note. Calculated by the author from a memorandum dated October 18, 1992,
Series 1. Personal/Political, Subseries Domestic Trip Files, box 29, f. 10, John
Glenn Papers, University Archives, The Ohio State University, Columbus, Ohio.
Table 5. Senator Barry Goldwater and going home, 1975–1980.
Note. Calculated by the author from daily and campaign schedules located in
Series V. Administrative, Subseries Travel, and Series I: Personal, Subseries
Senate Campaigns, Barry Goldwater Papers, Arizona Historical Foundation,
Arizona State University, Tempe, Arizona.
The Goldwater archives also contain the senator’s daily schedules and
itineraries that were prepared for his various trips back home to Arizona.
The record is unambiguous and exists in stark contrast to Glenn’s: Gold-
water did not travel home very often in the five years after his reelec-
tion in 1974, and when he did, he spent the bulk of his time in Mari-
copa or Pima County (see table 5). Maricopa County—essentially metro-
politan Phoenix—was Goldwater’s home county, and Pima County is the
home of Tucson’s second-largest city and only a ninety-minute drive from
Phoenix. Goldwater had good reasons for remaining in Washington: he
had had a series of operations on his hip and moving around was quite
difficult and painful, particularly in 1980, when a wire in his hip snapped,
requiring another operation that resulted in a secondary staph infection

(Goldberg 1995, 305; Goldwater 1988, 376).19 Goldwater also spent some
time traveling through the country building the Republican Party’s orga-
nization. Nevertheless, he did not visit his constituents in Arizona as often
or as widely as he may have in the past, and many of his trips centered
on speeches given to either supporters or high school graduates. In short,
the claim made by Bill Schulz that Goldwater had been scarce in Arizona
during the five years leading up to the 1980 election campaign was accu-
rate and true, confirmed by the office and campaign schedules found in
the archival record.
One might conjecture that Goldwater’s connection with his constituents

suffered because of his various ailments and out-of-state sojourns, which
made him vulnerable to the Schulz campaign’s aggressive effort. One
might also conjecture that John Glenn survived a tough reelection chal-
lenge despite the taint of the Keating Five scandals and the DeWine
campaign’s assault on his long-standing presidential campaign debt in part
because of his reconnecting with Ohio voters on his trips back home.
Of course, these accounts are not meant to provide a systematic test of
the electoral usefulness of going home, but if one collects enough travel
schedules across enough archival collections, a variable for the number
of days a candidate spends in the district can be created and inserted into
models predicting electoral success. I would hazard an educated guess that
building connections with constituents by traveling home helps members
of Congress get reelected—particularly when they are facing adverse
partisan demographics or difficult national electoral trends.
Voter�Mobilization�Matters
A campaign’s ground game—the grassroots effort to mobilize and target
voters—is nearly impossible to observe using standard campaign and elec-
tions data. Surveys such as the American National Election Study (ANES)
ask whether individuals are contacted by political parties or candidates
during a campaign. These self-reported contact questions are fraught with

difficulties. The most obvious is their dependence upon the respondent’s
memory recalling the contact. However, there is another problem from the
perspective of the campaign itself. The very nature of surveys precludes an
examination of a campaign’s grassroots strategy. Campaigns target partic-
ular voters in particular precincts while completely ignoring other voters
in other precincts. It is impossible to evaluate a campaign’s voter turnout
operation unless one has access to the very plan.
Often, in the files on campaigns, congressional archives will have the
tools necessary for the researcher to examine and independently eval-
uate a campaign’s voter targeting plan. In the case of Goldwater’s 1980
campaign, his longtime campaign manager Steven Shadegg had drawn up
a series of memos listing which precincts would be heavily targeted by
the campaign’s get-out-the-vote (GOTV) operation. Why is this impor-
tant? It is puzzling why Goldwater, in a heavily Republican year in a
Republican-leaning state, barely eked out a nine thousand–vote victory
in the general election with Ronald Reagan heading the ticket. Equally
puzzling is how the campaign held on despite a well-organized and well-
funded challenge mounted by a young and particularly energetic Democ-
ratic candidate. Those who worked for Goldwater suggest that the voter
turnout operation, especially the effort to encourage absentee voting in
Maricopa County, provided the key to victory.20 Assessing this claim
nearly thirty years later would be nigh impossible if it were not for the
Goldwater papers containing Shadegg’s strategy memo.
Shadegg’s memo divided precincts into four types: predominately

Republican precincts (R-A and R-B), swing precincts (S), and strong
Democrat precincts (D). He indicated that turnout efforts should focus
“first, on getting our supporters in the strongest Republican precincts to
the polls” and that “the Strong Democrat precincts (D) are not worth
any effort.”21 Attached to the memo are Pima and Maricopa County
precincts, grouped by these four designations. Pima County was a Demo-
cratic stronghold and home to Arizona’s other senator, Democrat Dennis
DeConcini. Maricopa County, in contrast, was Goldwater’s home county

and the supposed epicenter of a strong Republican GOTV effort headed
by the Maricopa County Republican Party.
Using these data presented in the Shadegg memo, combined with

precinct tallies available at the county recorders’ offices in both Pima
and Maricopa Counties, allows an assessment of the claim that Repub-
lican GOTV efforts helped to secure Goldwater’s victory. Although
similar information is not available from Bill Schulz’s campaign, Schulz
admitted in an interview that his absentee ballot and GOTV operation was
not as sophisticated as Goldwater’s.22 One can evaluate, first, whether
Goldwater’s vote totals improved upon Republican Sam Steiger’s vote
totals from the open-seat election in 1976, also a presidential election year.
Second, one can see whether voter turnout improved in 1980 relative to
1976. Finally, one can see the differences among areas that were specifi-
cally targeted by Goldwater and whether the Maricopa County operation
outperformed the Pima County operation. In short, it is possible to demon-
strate whether there is any empirical evidence supporting the speculative
notion that the GOTV effort in Maricopa County helped Goldwater hold
onto his seat.
Table 6. Comparison of Republican vote percentages in Republican, swing,

and Democratic precincts in Maricopa and Pima Counties, 1976–1980.
Note. Calculated by the author from precinct election returns for 1976 and 1980
provided by the Maricopa and Pima county recorders.
Table 7. Change in voter turnout between 1976 and 1980 in Republican,

swing, and Democratic precincts in Maricopa and Pima Counties
Note. See note in table 6.
Tables 6 and 7 appear to suggest that the answer is yes on all three
counts. Goldwater received a higher percentage of the vote than Steiger
did in 1976 in GOP-targeted areas—and received higher percentages
in Maricopa as compared to Pima County. But the real difference is
in voter turnout. In Maricopa County, turnout in Republican precincts
was up substantially compared to voter turnout in Democratic and swing
precincts. In Democratic Pima County, voter turnout in the strongest
Republican precincts was down. In the areas targeted by Republicans in
Maricopa County, more Republicans showed up in 1980 than in 1976,
and they gave a higher proportion of their vote to Goldwater than they
had done for the Republican nominee four years earlier. Finally, according
to voter registration and early voting statistics obtained from the Mari-
copa county recorder’s office, 4.7 percent of those voting cast ballots early
in 1980. This is up about half a percentage point from 1976 and is 1.7
percentage points higher than 1978.23 Considering that Goldwater lost
every other county in the state but one, the nearly 50,000-vote margin
in Maricopa County over Schulz likely pushed him over the top. And, it
would seem that the GOTV effort crafted by Shadegg and implemented by
the Republicans in Maricopa County helped keep Goldwater in the Senate
for another six years.
The�Party-Centered�Campaign�Did�Not�Disappear�in�the
Mid-Twentieth�Century
Scholars of political parties and Congress have long noted that the
mid-twentieth century was an important point of departure for the
conduct of congressional campaigns. Before the 1940s, most congres-
sional campaigns were conducted by party officials using party resources.
Indeed, many campaigns hardly featured the candidate at all: literature
focused mainly on party issues and platforms rather than on the positions
of candidates and their particular virtues as potential officeholders. But
beginning in the 1940s and 1950s, scholars began to take note of a new
campaign style: the candidate-centered campaign. These campaigns were
of a different breed; the candidate’s attributes and biography took center
stage, and issue positions were frequently tailored to fit the particular
interests and needs of the congressional district or state. Candidates began
to self-finance their campaigns, and using the new media of television and
radio, they spoke to voters directly in their homes rather than through the
filtered medium of the party and its precinct workers. It was generally
assumed that the transition between the two campaigns was quick, swift,
and even across the country.
Unfortunately, this conventional wisdom is simply not true. Party-

and candidate-centered campaigns coexisted during the mid-twentieth
century, and the type of campaign that was undertaken depended in large
part on the strength of the local party organizations within each state. In
areas with strong and vibrant party organizations that were well funded
and organized, candidates allowed the party to run the campaign. In areas
where strong partisan traditions never developed, the candidate called the
campaign shots and created personal organizations largely independent of
the party. I posited that congressional campaigns were more or less candi-
date-centered depending upon the local party’s ability to contribute the
resources necessary for the candidate to win election.
Using campaign correspondence, memoranda drafted by party and

campaign officials, television and radio advertising scripts, and campaign
brochures, congressional archives allowed a test of my resource theory

of campaigns. Sure enough, clear patterns began to emerge between
strong and weak party states. Illinois congressman Everett Dirksen’s
campaign against Senator Scott Lucas in 1950 demonstrated how different
a campaign in a strong party organization state looked when compared to
a campaign in a weak party organization state such as Idaho. Dirksen’s
campaign schedule was coordinated with parties at the state and local
levels, and as Dirksen indicated in correspondence with a supporter,
the campaign’s “organizational work would be handled by the county
chairman, the state central committees, the Cook County Committee, and
the Federation of Women’s Republican Clubs.”24 Fund-raising efforts
usually stressed the importance of electing a Republican ticket, not just
Dirksen to the Senate. Dirksen also appeared frequently with the entire
statewide Republican ticket on campaign trips around the state (Parker
2008, 94).
Compare this to Frank Church’s campaign for the Senate against Repub-
lican Herman Welker in 1956. According to a New York Times election
postmortem, Church “operated out of his own headquarters in Boise,
hardly bothering to check in with the Democratic State Headquarters”
during the course of the campaign.25 In campaign appearances throughout
the state, Church wisely avoided Stevenson and issues of national impor-
tance. Indeed, in a note penned by Church to national speakers visiting
Idaho in support of his campaign, he implored them to avoid these
issues in favor of Idaho’s economy and the record of his opponent.26
Church raised his own money and ran his own campaign with virtu-
ally no assistance from the weak Democratic organization in the state
or the national campaign committees. In short, Church ran the candi-
date-centered campaign that supposedly first emerged during the mid-
twentieth century. Dirksen, in contrast, hardly ran his campaign, deferring
to the party apparatus in the state. The archival record highlights these
important differences between the two campaigns for the researcher who
is looking for them.
The resource theory could not be tested with roll call data or elec-
tion returns. Only an examination of the campaign record provided by
congressional archives could tease out the relationship between individual
campaign organizations and the party organization. And, the story that
is told by these materials suggests that long-standing ideas about the
evolution of congressional campaigns were, at best, incomplete and more
complicated than existing accounts suggested.
Availability�of�Individual�Congressional�Campaigns’
Polls
Polls of individual congressional campaigns are widely available. Unfor-
tunately, one would not know it from much of the work that has been
done on congressional elections. There are three main resources scholars
utilize to explain congressional campaigns: the American National Elec-
tion Studies (ANES) time series; the 1978 ANES, which specifically
sampled 108 congressional districts; and the ANES 1988–1992 Pooled
Senate Study. The reason is obvious enough: it is exceedingly difficult
to obtain the monetary resources for individual researchers to conduct
frequent polls during individual congressional or Senate campaigns. The
problem with using national surveys to study individual congressional
campaigns is obvious: at most, there may be twenty or thirty respondents
interviewed in any one congressional district, and in any case, the survey’s
sample is designed to be nationally representative, not representative of a
particular state or district. Even the ANES Pooled Senate Study provides
a sample of only one hundred voters in each state—hardly enough to draw
any reasonable conclusions about the condition of a given Senate race. On
top of all of that, these surveys do not allow for a dynamic view of the
campaign. There is a presurvey and a postsurvey; that is it. The effects
of particular campaign strategies or important campaign events on voter
opinion cannot be measured, but only guessed at.
Fortunately, most congressional candidates—particularly incum-

bents—poll early and often. Frequently the researcher will find an exten-
sive benchmark survey that was undertaken on the incumbent’s behalf at
least a year or more before the official start of the reelection campaign.
Trend polls, at least two or three in the spring and summer months
during an election campaign, are also frequently found in the archival
record. And, if the researcher is particularly lucky, tracking data for the
last month of the campaign may also be present. New Mexico senator
Harrison “Jack” Schmitt’s reelection campaign was particularly prolific in
its polling efforts: not only were polls sponsored throughout his first term
and tracking polls conducted during the final six weeks of the campaign,
but the campaign even sponsored a postelection survey after Schmitt lost!
One limitation of these polls is the inability to conduct original analyses:

the raw data are generally not part of the collections. Instead, the polling
data in the archives are generally in the form of a research report drafted by
the pollster. Most of the time, the reports provide all of the original ques-
tions along with extensive crosstabs. Despite these limitations, the picture
that is presented to the researcher of an individual Senate campaign is
much richer than one relying upon national survey data, especially given
the propensity of political pollsters to ask open-ended questions about
incumbents and their challengers. For example, Harrison Schmitt’s loss
in 1982 is often described as a referendum on the Reagan administra-
tion and the Schmitt campaign’s negative tactics toward the end of the
campaign (Barone and Ujifusa 1983). But Schmitt had other problems
directly related to his representational connection that were not attrib-
utable to the Reagan administration and its policies. A survey done by
Tarrance and Associates for the Schmitt campaign in early September of
1982 asked those people who indicated they would vote against Schmitt
to explain why. Schmitt’s issue stances and his links to the Reagan admin-
istration account for 47 percent of the open-ended responses volunteered
by respondents. Nearly as many (44 percent) cited Schmitt’s personality
and legislative ineffectiveness.27 It is clear that although Schmitt’s close
connection to Reagan hurt him, he had not established a close represen-
tational connection with voters, such as the “one-of-us” relationship that

was advocated and enjoyed by New Mexico’s other Republican senator,
Pete Domenici. In general, a candidate’s polls shed light on resource allo-
cation strategies, the advertisements he or she ultimately chose to air, and
the places he or she chose to spend time on the campaign trail. They also
tell the story of how a member of Congress relates—successfully or not
—to his or her constituents.
Members�Communicate�Their�Washington�Work
Differently
In traveling home, members of Congress explicitly communicate their
representational style to their constituents. As Fenno (1978) noted,
members of Congress routinely employ three key styles: “one of us,”
policy expert, and constituent servant. It is hard, however, to observe these
styles using traditional congressional campaign data. Certainly, a review
of campaign advertisements is one way a researcher might proceed, and
the Wisconsin Advertising Project has these data—if the campaign took
place in 2000 or after and placed ads in one of the top one hundred media
markets.28 Before 2000, however, the researcher must rely upon news-
paper accounts and campaign advertisements located in archival collec-
tions. As a 2009 article in the American Political Science Review indicates,
however, campaign advertisements and media accounts are not always
unmediated, complete, or representative of the population of campaigns
(Druckman, Kifer, and Parkin 2009). In the same vein, newspaper articles
and campaign ads may not provide an unmediated, complete, or represen-
tative picture of a member’s representational style—particularly the style
he or she actively projects throughout the term in office.
The data that are available in congressional archives provide a second,

and most satisfactory, measurement of representational styles in the press
releases members themselves send to newspapers. Most congressional
collections have a nearly complete complement of press releases sent by
members of Congress throughout their careers, usually in the aptly named

press office files. These press releases are unmediated and are represen-
tative of how the member wishes to be portrayed publicly by the media.
Using the titles of the press releases, the researcher can code the repre-
sentational styles the member of Congress communicates to constituents.
Totaling up these communications can reveal which members stress posi-
tion-taking versus credit-claiming and whether particular patterns are
observed as a candidate moves closer to the election. In the case of
Alaska senator Mike Gravel, there was a discernible shift toward credit-
claiming about federal grant monies to communities as the election year
approached.29 This makes sense: Gravel represented a state that bene-
fited from a large influx of federal monies, and he specifically focused
his reelection campaign on how his seniority, experience, and influence
benefitted Alaska in Washington. Absent carefully observing a candidate
during campaign travel and his or her interactions with constituents, exam-
ining how press secretaries seek to portray the Washington work of their
members is perhaps the best way to observe the propensity of a member
to affect a “one of us,” policy expert, or constituent service reputation.
Lessons�Learned�and�How�to�Avoid�Repeating�Them
By now I hope I have convinced the reader of the theoretical and empirical
virtues of archival research. Archival research is a richly rewarding expe-
rience for the scholar of politics, and I find that it allows me to interact with
the political process in a very different fashion compared to my more tradi-
tional, quantitative research. Although the term is fraught with normative
implications, particularly among those people who study American poli-
tics, archival research allows one to establish a firmer grasp of the Amer-
ican campaign culture.
Other authors in this volume focus on the “how to” of archival research
and have provided valuable clues to the uninitiated. Rather than provide
the same tips, I offer instead three pitfalls the potential nonparticipant
observer should consider carefully before undertaking an archival trip.

They are access restrictions, a collection’s state of process, and for those
interested in campaign advertisements, the state of audiovisual materials
located within the collection.
The first consideration is whether a collection is open for research

or not. There are three types of access restrictions I have come across:
completely open, partially restricted, and closed. In the process of inves-
tigating a collection online, the archives will generally describe the terms
of access governing its use. If a collection is completely open, then the
researcher will have no problem; after establishing that the collection has
materials of interest that appear to be sufficient in quantity, he or she can
schedule a visit and go to work after thoroughly reviewing the collection’s
finding aid in consultation with the archivist responsible for the collection.
Partially or completely restricted collections, however, are another matter
altogether.
It is often the case that collections have sections that are closed to
researchers for a certain time period. Very often this includes material of a
particularly sensitive matter, such as case work requests pertaining to mili-
tary academy appointments, problems with social security checks (where
social security numbers may be listed), and the like. Patronage requests are
another area that is often closed to the researcher, usually until the death of
the donor. At other times, there may be no rhyme or reason for a restric-
tion. It is worth it to read these restrictions carefully. In one instance, I had
not and almost had to abandon a trip as a result: the materials I had hoped
to access about a particular campaign were closed despite the fact that
the senator of interest had been deceased for more than a decade. Gaining
access proved surprisingly difficult and took well over two months of e-
mail exchanges, letters forwarded to the family, and a copy of my book as
evidence of my academic qualifications. Access was finally granted, and
the trip was not cancelled. And, thankfully, the material I unearthed was
terrific and well worth my determined efforts.
In other cases, collections may be completely closed to researchers.

Do not be discouraged. It is always possible to request access to these
collections, and sometimes, the donor may be willing to allow a researcher
limited access. In these cases, it is important to become intimately familiar
with the collection’s finding aid and list exactly the boxes that are desired
for examination. It is helpful to work closely with the archivist on a letter
requesting access, because a carefully targeted request is more likely to be
successful than one that is sweeping in scope. At worst, the request will
not be granted. At best, access will be granted and the researcher may very
likely be the first to review the collection—other than the archivist who
initially processed it.
Access restrictions bring up another important point: oftentimes, collec-

tions that are restricted have yet to be processed by the institution holding
the collection. Dealing with an unprocessed collection should give the
researcher considerable pause; archival work is demanding when a collec-
tion has been well processed. Working with an unprocessed collection can
be like sorting through a grandparent’s attic: a long-lost family heirloom
may be found after shifting for hours and hours, or the same amount of
time might be spent looking at moldy back issues of National Geographic.
The whole expensive research trip may yield absolutely nothing at all.
I have worked with two unprocessed collections and one partially

unprocessed collection. In one instance, I had to open sealed boxes with
a letter opener. Most unprocessed collections have a rudimentary finding
aid that was put together by the congressional staff packing the boxes
as they were shipped off for storage. This can provide some direction to
the researcher. Second, the size of the collection relative to the member’s
career can provide some guide to the extent of the material that is avail-
able to the researcher. A member of Congress served for forty years and
has only two boxes in the collection? This is generally a bad sign. Third,
it is absolutely imperative to work with the archivist who is responsible
for the collection—assuming a researcher has been granted access—to
determine whether a trip is worthwhile. Even if a collection is completely
unprocessed, it is likely that someone has at least looked through it to get

some idea of what it contains. Finally, unprocessed collections simply take
more time to sift through. When a collection is well processed, certain
folders and even whole boxes can likely be excluded by their titles and the
content descriptions noted on the finding aids. In the case of unprocessed
collections, it is generally the case that every folder and box that looks to
be even remotely related to the research question will have to be looked
through. That said, unprocessed collections can be quite fun to work with
if one has the time and patience: they bring out the treasure hunter in every
researcher. Senator Harrison “Jack” Schmitt’s collection at the University
of New Mexico is one such unprocessed collection. I found actual raw
film footage on reels in six or seven boxes, representing the many hours
of footage media consultant Roger Ailes took of Senator Schmitt for use
in campaign commercials; I also found twelve different polls taken by the
Schmitt campaign beginning in 1978 to help plot strategy for the 1982
reelection campaign. The material provided a rich and detailed view of
the campaign and was well worth the effort taken to retrieve it.
Finally, a note for those scholars who are interested in reviewing

campaign advertisements, either radio or television. One of the most frus-
trating aspects of doing archival work is hitting the proverbial research
gold mine and then not being able to reach the gold. This is certainly the
case at times with audiovisual materials. Television advertisements were
often put onto Beta or ¾-inch tapes, and radio ads were often recorded
on reels. Unfortunately, it can be incredibly frustrating as a researcher
to access what is on these tapes. Three barriers exist. First, the deposi-
tory may not have the proper audiovisual equipment to view these mate-
rials in-house. In fact, this is often the case. Second, the depository may
be reluctant to allow a researcher to view the materials because of their
age and fragility. This is unfortunate given that these materials provide
an important and rich record of the campaign that may not be accessible
anywhere else.30 Third, some collections are more than willing to let a
researcher review the materials if he or she will pay to have them trans-
ferred to electronic media, such as a DVD or CD. This sometimes involves
paying for two copies: one for the researcher to take home and one for
the archives to keep. Sometimes this is relatively inexpensive, but in one
instance, I was asked to pay more than $250 to transfer four audio-tape
reels (I politely declined). The researcher should be aware of these costs
and work them into any budget requests in advance if audiovisual mate-
rials will be viewed.
Conclusion
It is worth ending on a note of caution for the eager researcher who is
ready to dive into the archives. Ultimately, political scientists often want
to draw generalizable conclusions from their research. It is critical, there-
fore, to consider the very real threats of bias that can creep into one’s
research design if one is not vigilant. Two bias threats are particularly
problematic. The first is selection bias, which can be mitigated by careful
research design and the proper selection of cases. Unlike large n-quantita-
tive studies relying upon the rules of statistical inference, cases should be
purposively and not randomly selected. It is important to be sure, however,
that the selection rule is not correlated with the dependent variable of
interest. Otherwise, the causal relationships observed between the depen-
dent variable and the independent variables may actually be driven by the
selection rule—which represents a possible omitted variable that may be
the real causal mechanism driving the phenomenon under study. Those
scholars who engage in archival work can be too quickly dismissed as
simple storytellers passing on interesting, but perhaps not representative,
anecdotes. To make archival research truly compelling and informative,
one must take selection bias seriously and avoid it as much as possible
when constructing research design. Importantly, this means allowing “for
the possibility of at least some variation on the dependent variable” of
interest (King, Keohane, and Verba 1994, 129). It also means spending
time at less-than-desirable locations instead of jetting off to Hawaii or
Colorado whenever one wants to study at a congressional archive.
The other problem is that the historical record may itself be biased.
Some documents are more likely to have survived to the present day,
whereas some actors may be less likely to have left written or docu-
mentary accounts of their actions behind (Lustick 1996). In the case of
senators and members of Congress, this is usually not problematic: their
activities are often well documented by them and their staff. The issue
can become troubling, particularly for those scholars studying campaigns,
with the records of unsuccessful challengers. Rarely are the records of
those who fail to get elected saved and archived.31 This is less problem-
atic when studying Senate campaigns simply because few first-time candi-
dates run for the Senate. Many Senate challengers are politically experi-
enced, having enjoyed extensive public careers that generate vast research
collections. But for members of the House, the problem is much more
widespread. In short, history better documents those who win elections
and serve in office than those who lose, and this implication is important
to remember before drawing any conclusions about the campaign efforts
of failed challenger campaigns, particularly in relation to the incumbent
candidate.
Archival research holds great potential for the political scientist. The
possibilities for new avenues of research are limited only by the time,
funds, and tenure status of the researcher. The information that is available
in congressional archives sheds new light on old questions, provides new
measurements of theoretical concepts, and helps build new theories. Most
importantly, archival records help political scientists to rediscover polit-
ical process, which is often messier and less certain that theoretical models
sometimes suggest. This is not an indictment of the discipline or quanti-
tative research more generally, but merely an issue of perspective: some-
times it is worth observing politics up close and in detail before pulling
back and looking at the bigger picture. Induction and deduction are two
sides of the same research coin, and it is worth remembering that for all the
attention political scientists pay to deduction, induction is another useful
tool in their research kits that can also powerfully illuminate the political
world.
Endnotes
1. I acknowledge the generous financial support of Montana State Univer-

sity in the form of a scholarship and creativity grant and research start-
up funds provided by the vice president for research, creativity, and tech-
nology transfer. I would like to thank the archivists at the many research
collections I have visited over the years for their expert assistance and
advice. Three stand out for going above and beyond the call of duty: Linda
Whitaker of the Arizona Historical Foundation, Jeff Thomas of the John
Glenn Archives, and Alan Virta of the Special Collections at Boise State
University. Their extensive knowledge of their collections has made my
work easier and more enjoyable and, I hope, has allowed me to write better
and more interesting books and articles. I am grateful especially to Sean
Kelly for asking me to write this chapter and for providing so much schol-
arly advice and friendship.
2. See also Frisch and Kelly (2003) for an explication of the similar chal-
lenges faced in doing archival research.
3. For a discussion of the method of archival research more generally and
its value to political scientists and the study of Congress in particular, see
Harris (2005).
4. Kelly (2004) echoed some of the pitfalls and drawbacks of participant
observation when discussing his experience as an APSA congressional
fellow.
5. Memo to Jack Schmitt from Pete Domenici: “The Following Observations
and recommendations are based upon personal contact with the Schmitt
Campaign and Personnel, analysis of media impressions of the campaign,
discussions with other politically interested persons, and certain very
fundamental principles not of politics, but of human nature,” n.d., f.
“Domenici Memo,” box 634, Harrison “Jack” Schmitt Papers, Center for
Southwest Research, University of New Mexico. The Schmitt papers are
unprocessed; the box number here and in other references is the original
number assigned by Schmitt’s staff upon transferring his papers to the
University of New Mexico.
6. Ron Crawford is one such example. He worked out of Washington,
DC, and was Goldwater’s fund-raiser. His correspondence can be found
throughout the Goldwater papers, but he is not someone who became inti-
mately involved in the day-to-day operation of the campaign until the fall
of 1980 (Goldberg 1995, 308). Crawford is an elusive fellow, and I spent

some time tracking him down for an interview, locating him only through
a chance e-mail sent to his wife after an extensive Internet search.
7. Memorandum, Harrison Schmitt to District Staff, “What has he done?”
December 3, 1979, box 635, Harrison “Jack” Schmitt Papers, Center for
Southwest Research, University of New Mexico.
8. W-NOMINATE scores avoid some of these problems by using only the
votes of a single Congress.
9. A new film produced in part by C. C. Goldwater, Senator Goldwater’s
granddaughter, paints a more nuanced picture, particularly on the issue of
gay rights and abortion. See Mr. Conservative: Goldwater on Goldwater
(2006).
10. The entry “Planned Parenthood Cocktails” appears on Goldwater’s
calendar, located in Series V: Administrative, Subseries: Administrative,
96th Congress, box 534, f. 16, Barry Goldwater Papers. Entries for April
15, 1977, and October 1979 also list Planned Parenthood events in the
Goldwater home (see box 534, f. 15 and box 530, f. 12 in the same series).
11. “KOOL-TV/Arizona Republic Meet the Candidate’s Forum,” Series 3:
Media Files, Box 7, f. 10 William S. Schulz Papers, Arizona Histor-
ical Foundation, Arizona State University, Tempe, AZ. In fact, Gold-
berg noted that after Goldwater left the Senate in 1987, he was “no
longer constrained by political considerations…and championed abortion
rights…[by] agree[ing] to serve on [the National Coalition for Choice]
national advisory board” (1995, 331).
12. Schulz television ad “Flips,” Series 3: Media Files, Box 7, William S.
Schulz Papers. See also Goldberg 1995, 308.
13. Of course, Poole and Rosenthal (1997) cautioned readers about making
these types of errors when utilizing DW-NOMINATE estimates.
14. Carl E. Anderson to Herman Talmadge, October 25, 1977, Series II: Press
Office files, 1957–1980, Subseries D: 1977, box 80, folder 10, Herman
Talmadge Papers, Richard Russell Library for Political Research and
Studies, University of Georgia Libraries, Athens, Georgia. Emphasis is in
the original. I calculated the tally of constituent opinion from constituent
correspondence contained in folder 10 and boxes 56 to 58 in Subseries C:
1978 of Series II: Press Office files.
15. The famous ad “Bloodhounds” can be viewed on YouTube: http://www.
youtube.com/watch?v = bcpuhiIDx3Q.
16. Television ad labeled “Duncan,” airing October 10, 1980, Box 7, William
Schulz Papers, Arizona Historical Foundation, Arizona State University,
Tempe, Arizona.
17. Memorandum for the Record, “Reflections,” April 27, 1979, Series I:
Personal, Box 26, f. 29, Barry Goldwater Papers.
18. See Parker and Goodman 2009, Parker and Goodman n.d., and Goodman
and Parker 2010 for a discussion of the problems and challenges in
utilizing House and Senate office expenditure data.
19. Judy Eisenhower, phone interview with author, July 24, 2009.
20. Ibid.; Ronald Crawford, phone interview with author, June 26, 2008.
21. Letter from Stephen Shadegg to Mrs. Tom Fannin and Peter Dunn,
September 19, 1980, Stephen Shadegg Papers, Arizona Historical Foun-
dation, Arizona State University, Tempe, Arizona (unprocessed).
22. William R. Schulz, personal interview with author, June 5, 2008.
23. Information provided to the author by the Maricopa county recorder’s
office in the form of an Excel spreadsheet. Data are available from the
author upon request.
24. Dirksen to Fred A. Burt, June 1, 1950, f. 150, Everett M. Dirksen Papers,
Dirksen Congressional Center, Pekin, Illinois. Quoted in Parker 2008, 95.
25. “Church of Idaho Noted as Speaker,” quoted in Parker 2008, 82.
26. Memorandum. “Concerning Certain Issues of Local Importance to Idaho
in the Coming Election Campaign,” n.d. series 5.1, box 2, f. 12, Frank
Church Papers, Special Collections Library, Boise State University.
Quoted in Parker 2008, 81.
27. Tarrance and Associates, September 1982, “A Survey of Voter Attitudes
in the State of New Mexico,” box 604, Harrison “Jack” Schmitt Papers,
Center for Southwest Research, University of New Mexico.
28. The Wisconsin Advertising Project is at http://www.polisci.wisc.edu/
~tvadvertising.
29. I examined and coded the titles of all press releases sent by Gravel’s Senate
office between 1975 and the Democratic primary in August of 1980.
The press releases can be found in Series XXXI: Press Releases, 1969–
1980, Mike Gravel Papers, Arctic and Polar Collections, Elmer Rasmuson
Library, University of Alaska.
30. The University of Oklahoma hosts the Julian P. Kanter Political Commer-
cial Archive, which holds perhaps the most comprehensive collection of
American political advertisements in the country. Many ads from presi-
dential, Senate, and House campaigns are located here, and there are many
commercials that are not located in individual congressional archives.
31. One notable exception is Bill Schulz’s collection at the Arizona Historical
Foundation. I am grateful that Schulz retained the records of his unsuc-
cessful Senate bid and donated them to the foundation.
Bibliography
Andersen, Kurt, Christopher Ogden, and B. J. Phillips. 1984. “The Senate:

Landslide or No, the G.O.P. Margin Shrinks.” Time Magazine, November
19, 94–97.
Bachrach, Peter, and Morton Baratz. 1962. “Two Faces of Power.” American
Barone, Michael, and Grant Ujifusa. 1983. The Almanac of American Politics,
1984. Washington, DC: National Journal.
Church, Frank F. Papers. Special Collections Library, Boise State University,
Boise, Idaho.
“Church of Idaho Noted as Speaker.” 1956. The New York Times, November
7, 17.
Coleman, John J., and Paul F. Manna. 2000. “Congressional Campaign
Spending and the Quality of Democracy.” Journal of Politics 62: 757–789.
Dirksen, Everett M. Papers. Dirksen Congressional Center, Pekin, Illinois.
Druckman, James N., Martin J. Kifer, and Michael Parkin. 2009. “Campaign
Communications in Congressional Elections.” American Political Science
Review 103: 343–365.
Edwards, Lee. 1995. Goldwater: The Man Who Made a Revolution. Wash-
ington, DC: Regency.
Fenno, Richard F. Jr. 1978. Home Style: House Members in their Districts.
New York: Longman.
———. 2007. Congressional Travels: Places, Connections, and Authenticity.
New York: Pearson Longman.
Frisch, Scott A., and Sean Q Kelly. 2003. “Don’t Have the Data? Make Them
Up! Congressional Archives as Untapped Data Sources.” PS: Political
Science & Politics 36(2): 221–224.
———. 2006. Committee Assignment Politics in the U.S. House of Represen-
tatives. Norman: University of Oklahoma Press.
———. 2008. Jimmy Carter and the Water Wars: Presidential Influence and
the Politics of Pork. Amherst, NY: Cambria Press.
Glenn, John. Papers. University Archives, The Ohio State University,

Columbus, Ohio.
Goldberg, Robert A. 1995. Barry Goldwater. New Haven, CT: Yale Univer-
sity Press.
Goldwater, Barry M. 1988. Goldwater. New York: St. Martin’s Press.
———. Papers. Arizona Historical Foundation, Arizona State University,
Tempe, Arizona.
“Goldwater Creates Confusion with Backing of Abortion Ban.” 1980. Arizona
Republic, October 22.
Goodman, Craig, and David C.W. Parker. 2010. “Who Franks? Explaining the
Allocation of Official Resources.” Congress & the Presidency 37: 254–
278.
Gravel, Mike. Papers. Alaska and Polar Region Collections, Elmer E.
Rasmuson Library, University of Alaska, Fairbanks, Alaska.
Harris, Douglas B. 2005. “Recovering History and Discovering Data in the
Archives: An Alternative ‘Mode of Research’ for Congress Scholars.”
Extension of Remarks 28. http://www.apsanet.org/~lss/Newsletter/july05
/EOR-July-2005.pdf.
Hill, Kim Q., and Patricia A. Hurley. 2002. “Symbolic Speeches in the U.S.
Senate and Their Representational Implications.” Journal of Politics 64:
219–231.
Kelly, Sean Q. 2004. “Shaping Congressional Studies: The APSA Congres-
sional Fellowship Program at 50: Editor’s Introduction.” Extension of
Remarks 27: 1–5. http://www.apsanet.org/~lss/Newsletter/jan04/January0
4-EOR-Complete.pdf.
King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing Social
Inquiry. Princeton, NJ: Princeton University Press.
Lustick, Ian S. 1996. “History, Historiography, and Political Science: Multiple
Historical Records and the Problem of Selection Bias.” American Political
Matthews, Jeff J. 1997. “To Defeat a Maverick: The Goldwater Candidacy
Revisited, 1963–1964.” Presidential Studies Quarterly 27: 662–678.
Mr. Conservative: Goldwater on Goldwater. 2006. Julie Anderson, director.
Zeitgeist Films.
Parker, David C. W., 2008. The Power of Money in Congressional Campaigns,

1880–2006. Norman: University of Oklahoma Press.
Parker, David C. W. and Craig Goodman. 2009. “Making a Good Impression:
Resource Allocation, Home Styles, and Washington Work.” Legislative
———. n.d. “Our State’s Never Had a Better Friend: Resource Allocation
and Home Styles in the Senate.” Manuscript under review.
Poole, Keith T., and Howard Rosenthal. 1997. Congress: A Polit-
ical-Economic History of Roll Call Voting. New York: Oxford University
Press.
Schmitt, Harrison “Jack.” Papers. Center for Southwest Research, University
of New Mexico, Albuquerque, New Mexico. Unprocessed.
Schulz, William R. Papers. Arizona Historical Foundation, Arizona State
University, Tempe, Arizona.
Shadegg, Stephen. Papers. Arizona Historical Foundation, Arizona State
University, Tempe, Arizona. Unprocessed.
Talmadge, Herman E. Papers. Richard B. Russell Library for Political
Research and Studies, University of Georgia Libraries, Athens, Georgia.
Yiannakis, Diana Evans. 1982. “House Members’ Communication Styles:
Newsletters and Press Releases.” Journal of Politics 44: 1049–1071.
Chapter 6
The�Search�for�the
Elusive�Executive
Archival�Data�Collection
Methods�at�Presidential�Libraries
Brandon Rottinghaus
Democratic thinking assumes the following: as the only national public

representative, presidents should follow the preferences of the public on
policy matters. However, presidents are graced with a national stage and a
larger-than-life public presence, a fact that compels, perhaps even forces
them to be active in the use of the “bully pulpit” to lead public opinion. So,
do presidents lead or follow public opinion? When considering whether
or not to lead or follow, do they consult public opinion surveys to inform
them about what the public may or may not support? Unraveling the causal
mechanisms behind such a contradictory set of conditions (leading and/or
following public opinion) presents serious methodological trouble to even
the most intrepid social scientist. Scholars have explored this question
using heavily quantitative analysis with success, but the question about the
actual internal White House use of public opinion data can be answered
only by examining the historical record from inside the White House. In
this case, what is needed is primary evidence that documents the ways
in which the president and his staff use public opinion to follow or lead
public opinion.
After having grappled with this question, and having visited each free-
standing presidential library to collect evidence to address this puzzle,
I found that the documentary archival data demonstrated several inter-
esting and telling trends concerning the White House’s use of public
opinion data (see Rottinghaus 2008, 2010). In particular, the archival
record illustrated that presidential advisers were worried about appealing
to public opinion (making them appear to pander on some issues), public
opinion was used to pressure Congress when public opinion was on the
president’s side, the White House would attempt to manipulate public
opinion with polls (and through the interpretation of these polls), presi-
dents and their staffs were concerned about partisan publics as much as the
mass public, and most importantly, presidents have the ability to use the
“bully pulpit” to communicate with the public and effectively change their
minds. In using the document collections at several presidential libraries,
I more clearly understood how the White House worked to manage public
opinion, including the assumptions that were made, the strategies that
were attempted, and the outcomes that were understood.
In this chapter, drawing on my experience at multiple presidential

archives over several years, I describe the lay of the land in terms of
the presidential archival system and provide an overview of the research
process at these libraries, including where to research, how research is
conducted, and potential pitfalls that may hinder archival research or the
interpretation of the archival data that are collected. My goal is to give
readers a general overview and a sense for how my research needs were
met with the methods I used and the techniques I employed.
The Search for the Elusive Executive 205
Essential�Considerations
The papers of the presidents who served before President Herbert Hoover
are maintained largely by the Library of Congress (LOC) in Washington,
DC. The LOC has most of the documents from these early occupants of
the Oval Office, but records are also scattered in other local collections
(such as universities or historical societies). The Library of Congress’s
website notes:
The Manuscript Division of the Library of Congress is the nation’s

oldest and most comprehensive presidential library, for although
the recently built presidential libraries each hold the papers of a
single chief executive, the Manuscript Division has in its custody
the papers of twenty-three presidents, including the men who
founded the nation, wrote its fundamental documents, and led it
through the greatest crisis of its existence.1
Table 8 lists the individual presidents and the dates of the collections in
the LOC. These documents are generally less voluminous than those for
later presidents and are available from the main reading room in the LOC.
Table 8. Library of Congress presidential manuscript collections.
Note. Table adapted from the Library of Congress Presidential Manuscript Divi-
sion website.
For presidents serving after Calvin Coolidge, the National Archives

and Records Administration (NARA) administers the presidential library
system, which includes individual stand-alone presidential libraries for
Presidents Hoover to (as of the time of this writing) George W. Bush
—future presidents will also have similar stand-alone libraries. These
libraries are governed by the Presidential Library Acts of 1955 and
1986, the Presidential Records Act of 1978, the Presidential Recording
and Materials Preservation Act of 1974, and various executive orders.2
Table 9 lists the locations and websites of each of the freestanding pres-
idential libraries. Some of the documents pertaining to the Nixon presi-
dency continue to be located in the National Archives II in College Park,
Maryland (although most of the records were moved to California in
February of 2010); for instance, the Oval Office recordings during Presi-
dent Richard Nixon’s time in office (January to July of 1973) will remain
in the National Archives.
Table 9. NARA presidential library websites and locations.
Organization�of�the�Archives
All presidential archives are arranged a little differently, but most have
similar characteristics. The following section describes the general orga-
nization of the archives, with an emphasis on the stand-alone presidential
libraries. Keep in mind that a quality search should start with the finding
aid and examine the papers of department offices, specific subjects, key
individuals serving as administrative secretaries, press department offi-
cials, policy advisors, and/or department secretaries. General papers of

the president are also available for searching, including the White House
central files (including subject file, name file, and chronological file), the
White House official files, the president’s personal files, and (for some
administrations) the president’s secretary’s files. Papers related to cabinet-
level departments, executive agencies, regulatory agencies, and presiden-
tial staff are also available. Each of these sets of files is important to under-
stand individually and is discussed in the following text.
The presidential archives of Franklin Roosevelt, Harry Truman, and
Dwight Eisenhower have internal organizations (remaining from the old
White House filing system) which are different from the archives subse-
quent to John Kennedy. The White House filing system pre-Kennedy was
staffed by permanent civil servants (some of whose tenure began under
the administration of William McKinley). This filing system possessed
no true centralized or numerical system (relying on an antiquated and
anachronistic U.S. Army filing system), which makes research trickier.
Finding aids are available at each library which show the organization
of the papers at the library; these documents are updated periodically
and are often available online. For the archives of President Kennedy to
the present, individual documents are usually cross-referenced among the
papers to provide continuity (as will be explained in detail subsequently).
The White House central files in the Roosevelt, Truman, and Eisen-
hower collections are divided into the official file, the general file, the
confidential file, and the president’s personal file (or the president’s
secretary’s file). The White House official file contains files with a
substantial quantity of factual materials on major issues of public policy,
including executive orders, records of press conferences, and bills or reso-
lutions that have been signed. These are usually just copies of letters sent
in acknowledgment of other letters or reports sent to the president or
additional hard copies of White House administrative materials, usually
from important or influential citizens. The general files contain letters
from individual citizens, interest group organizations, and occasionally
members of Congress on matters of interest. These documents are usually

simple acknowledgments from the White House staff and give no internal
insight on presidential thinking. The White House confidential files are
correspondences, memoranda, telegrams, and reports (often once-secu-
rity-classified material) which originated from or were sent to senior
White House staff or the president.
The White House central file subject file (commonly referred to as

the White House Office of Records Management subject file) contains
an alphabetical listing of documents sorted by subject. For example,
according to the George H. W. Bush Library website:
The White House Office of Records Management (WHORM)

Subject File consists of 58 broad subject classifications developed
by the Office of Records Management during the Kennedy Admin-
istration to facilitate the filing and retrieval of White House records.
The WHORM system allows White House personnel to file indi-
vidual documents according to subject. Therefore, the Subject File
is comprised of a limited number of main (or primary) categories
(in this case 58) which in turn are divided into sub (or secondary)
categories. Each primary category is assigned a two letter code,
i.e. FI for Finance. All subcategories listed under finance are also
assigned the same two letter code (FI) with a numeric extension
of up to six numbers. For example: FI001 stands for Accounting–
Audits while FI004-02 stands for Estimates–Budget. Within each
category or subcategory, each document is assigned a unique six
digit number.3
Three other kinds of files are frequently organized under the umbrella
of the central files. First, the White House central files also often contain
a name file. From the Ford Library:
The WHCF Name File is a name index to the Subject File. Cross-
references to the Subject File are filed under the names of Sena-
tors and Representatives; organizations, corporations, and institu-
tions; local, state, and foreign government officials; businessper-
sons, educators, celebrities, and office seekers; and many private
citizens. A few projects are also indexed, at least partially, e.g.

“Swine Flu Program.” The Name File also includes many items
deemed too unimportant for the Subject File. The Name File has
only marginal value for names of White House staff and certain
other Executive branch officials.4
For my research, the documents in the name file were an easy way to
locate the documents linked to a particular individual, such as a White
House staff member or a member of Congress, who I suspected may have
provided the president with information on public opinion.
Second, the central files also often contain a related chronological file.
These documents are chronologically arranged carbon copies of letters
sent over the president's facsimile signature (or sometimes a real signa-
ture). If a researcher were interested in determining the executive history
of presidential activity, the chronological file can help him or her piece
together the time trends. For instance, did the president know that people
supported a policy idea before he advocated that policy? Such informa-
tion is critical to identifying the nature of the president’s action as leader-
ship or responsiveness to public opinion. Third, the central files frequently
contain a social file, which contains documents related to the social activi-
ties of the presidency. These files often include material on the first family,
state dinners, presidential vacations, and other presidential trips.
The White House central files in presidential archives, including and

subsequent to John Kennedy, are divided into two types of files: general
and executive files. General files are those handled by lower-level White
House staff members and include memoranda, letters, and reports from
individual citizens or “unimportant” interest groups. Executive files are
similar memoranda, letters, and reports, except that these forms are from
higher-profile (or “important”) groups such as members of Congress,
former presidents, influential citizens, and members of the media, and
these letters were written by or to senior White House staff. It is therefore
more appropriate to view the latter files with greater concern and more
intensity because it is in these files where most of the important policy
decisions were made. Locating particular important documents in these

archives is easier because every document is cross-referenced back to the
central files holdings. For instance, if a specific memorandum is known to
be in the files of the attorney general, that memorandum is cross-listed to
the White House central files folder on that issue (but only if the memo-
randum is known to be located in that file). This is in contrast to the pre-
Kennedy filing system, where documents were not centrally classified
by subject and could not be cross-referenced. This often makes checking
the provenance and history of a document more difficult. For instance,
tracking which advisers viewed a particular memorandum in an attempt
to determine who knew what and when is more difficult if the location of
each memorandum is not uniformly identified in each location.
Besides the central files, other specific offices within the West Wing or
East Wing of the White House may have particular files, such as the Office
of Presidential Speechwriting, the Office of Congressional Liaison (this
office changes names across the administrations), the Office of Chief of
Staff, the White House Press Office, the White House Counsel’s Office,
the White House Social Office, the Office of the First Lady, the Council
of Economic Advisers, and the President’s Personnel Office. These indi-
vidual offices are sometimes large and well established, such as the White
House Counsel’s Office or the Office of the Chief of Staff. They are
often unique to a particular presidential administration’s policy or political
needs. For instance, in the Jimmy Carter Library, the finding aid for the
counselor to the president on aging reads:
Nelson Cruikshank served as Counselor to the President on Aging

from January 1977, until April 1980, when he was replaced by
Harold Sheppard. Richard Conn, a detailee from the Department of
Labor, served as assistant to both. The Counselor on Aging served
as the principal White House adviser on all issues dealing with the
elderly and was the White House liaison with the Federal Commis-
sion on Aging.5
The files for individual members of senior White House staff (special
assistants to the president, special counsel to the president, or counselor
to the president) are also available for research.
Similarly, documents related to cabinet-level offices, such as the U.S.

Department of State, U.S. Department of Energy, U.S. Department of
Housing and Urban Development, U.S. Department of Defense, and U.S.
Department of Commerce, are housed at the archives of the presidents
who were charged with their management. The documents from other
regulatory or executive agencies are also often located at these libraries,
including the U.S. Agency for International Development, Bureau of the
Budget (later the Office of Management and Budget), U.S. Commission
on Fine Arts, U.S. Commission Civil Rights, Federal Communications
Commission, Federal Trade Commission, and General Services Admin-
istration. As the size of government grew over time, the umbrella of
presidential documents from these executive agencies and departments
expanded. This allows researchers access to an unprecedented wealth of
information, although the volume can be overwhelming.
Similarly, many archives separate files which are more personal in

nature to the president’s daily business while he is in office. For instance,
the president’s personal file (most frequently present for older presidents,
such as Hoover, Roosevelt, and Truman) contains a president’s social
correspondences, congratulatory letters, and other memos to and from
the White House staff. In some archives, these papers are called the
president’s secretary’s files and include private documents initiated by,
received by, or kept by the president’s personal secretary. These include
documents the president wrote on (the originals are frequently kept sepa-
rate from duplicates in a “president’s handwriting file”), materials from
the president’s outbox, and other documents related to the president’s daily
“homework.” It is in these files where one can get a window into the
inner workings of the White House staff and often an intimate view of
the president’s viewpoints on a range of issues. Postpresidential papers
are generally also a part of the textual record at most libraries, especially
for those presidents who lived for a long period of time after leaving the
office.
Of course, certain archives have particular sets of papers or files that
are unique to the administration or endemic to specific individuals within
the auspices of the president in question. These documents can also be a
valuable source of information because they are specific to an individual,
event, or organization. For instance, President Eisenhower’s personal
secretary, Ann Whitman, created several chronological files containing
items the president saw that week or documents he requested. Similarly,
at several libraries, the diaries of specific individuals are valuable for the
insight into the politics, personnel, and issues of the time. Also, papers of
the political parties (and often key individuals), such as the Democratic
National Committee or the Republican National Committee, involved
during the term of the president’s service are commonly a wealth of quality
data. These topics often contain records of the president’s campaigns for
public office; these are not generally the property of the government,
although they are donated to the archive, so special permission may be
required to access them. These organizations or activities were frequently
headed by staff who were close to the president, making the documents
in them relevant for analysis. For instance, much of the public opinion
polling done by the White House’s reelection efforts is integrated into the
daily presidential activities, making these documents of particular impor-
tance to my research for identifying how presidents and their staffs were
influenced by public opinion polling.
Typically, a presidential archive will also have two other kinds of non-
document-related collections. First, all archives have audio and visual
materials that can be accessed in the same manner as the textual materials.
These materials usually include audio recordings, video recordings, still
photographs, news interviews, press conferences, presidential speeches,
and other materials. A researcher who is interested in working with these
materials should consult with the library’s audiovisual archivist (which
each presidential library employs). Second, most libraries (especially
those that have been in place longer) have oral histories of staff, family,
friends, business associates, government workers, presidential staff, and
other notable figures who knew or worked with the president. These oral
histories tend to be taken shortly after a president leaves office but are
also taken periodically by library staff or at specific conferences. Texts of
these interviews are usually requested in the same manner as other textual
documents. These interviews are often very candid and give the researcher
a clearer sense of an issue or event. In my work, advisers who were timid
about a written endorsement of the use of public opinion as a reason for
action were much more forthcoming when they were asked about it in
these oral history interviews.
Beyond the official federal records of a president’s administration (the

documents that have been described in the preceding paragraphs), all
presidential archives contain some mix of the personal papers of many
cabinet members, business associates, staff members, friends of the pres-
ident, influential advisors, writers and academics of the time, and family
members. These documents (usually indicated by “files” or “papers”
instead of “records”) are typically donated to the presidential archive
subsequent to the president’s term in office. These papers are also a valu-
able source of information on issues or events that may contain a different
or value-added perspective. For my work, the papers of Robert “Bob”
Teeter, who served as the pollster for Presidents Gerald Ford and George
H. W. Bush, were useful as a source of his style of reporting the White
House’s internal polling data to the president and his staff. Often, however,
these documents are simply duplicates of documents that are contained in
the official records (several copies of each official documents are made)
and are typically not cross-referenced to the original document in the offi-
cial White House files. These papers are frequently subject to restrictive
deeds of gift and may require special permission to access (this will be
described in more detail in the following section). These papers are also
often only partially processed, depending on the size of the collection and
the recency of their acquisition.
Gaining�Access
These records are government documents, so they are publically acces-
sible to anyone who wants to see them. The first step for a researcher is
to contact (via e-mail or phone) the library that he or she wants to visit
to ask about the collections or topics of interest. At that point, an appoint-
ment can be made; this is not required but is recommended so that the
archival staff can have the materials ready when the researcher arrives.
Upon arrival at the library, the researcher will be asked to show govern-
ment-issued identification and fill out a short form, including information
about his or her researcher status (academic, public, student), the scope
of the research (if applicable), and the general topics he or she desires to
research. Information about the research agenda is not required to gain
access to the papers, but this information will help an archivist to make
the researcher’s visit more productive because he or she will know what
is being sought. A research card with a unique number will be assigned;
it must be used to sign in each time a researcher visits the research room
of a library or when requesting documents from the archivists. This card
expires one year after the date it is assigned but can be updated at any time.
The first step of a research strategy should be to consult the finding

aids, which should be available online and will certainly be available at the
library. These catalogues are a research road map. If research is being done
online, it is important to ask the archivists if there are any updates that have
not been put online yet. The archival staff may update and distribute these
in the research room before they are put online. Likewise, if the archival
search process is started online, a researcher can consult the Archival
Research Catalog (ARC) online to start the search. These records tend to
be less well updated than the finding aids either in the research room or
on the website for each library, so they should be used as a starting point
rather than an ending point.
If the documents related to a subject, person, or event are not currently

available for research because these records have not been processed
by the archive, a researcher can file a Freedom of Information Act
(FOIA) request.6 A FOIA request will formally begin the process to

open the records in question and request that the archive begin system-
atic processing of the papers related to the FOIA request. A researcher
should consider filing a FOIA request if the documents related to a subject,
person, or event he or she desires to see are not currently available for
research because the archivists have not yet processed the information (a
separate process for requesting to see classified materials is discussed in
the next paragraph). FOIA requests are valuable to researchers because
they can give access to the new documents; however, these requests are
processed in order, so it may be several years before a researcher is
allowed access to those documents. Documents processed by a FOIA
request are eventually integrated into the permanent papers, but they are a
good way to search for specific issues, people, or events before the docu-
ments are processed in order. In more recent presidential libraries, FOIA-
based documents are often listed separately from the permanent collec-
tion. For instance, at the George H. W. Bush and Bill Clinton Libraries
(as of this writing), each archive has few total documents processed, so
it makes sense to start with those documents that have been processed by
FOIA request. Likewise, archivists regularly update previously classified
documents, so it is helpful to inquire about these possible collections.
A researcher can also specifically make a request for individual docu-

ments that have been processed but are classified (or restricted) for reasons
of national security to be unclassified while he or she is at the library;
these can be done on a case-by-case basis.7 Some can be done adminis-
tratively (or internally by the archival staff or the head archivist of the
library), whereas some require a more lengthy external process of declas-
sification by national security personnel. Summing this up, the National
Archives outlines the following procedures for gaining access to national
security–classified records:
(1) If a Presidential record or a portion thereof is closed under a

restriction of the Presidential Records Act (PRA), with the excep-
tion of national security classified information, the closure may be

appealed by the researcher.
(2) An appeal should be filed in writing no later than ten working

days after the researcher receives written notification that access to
the Presidential records has been denied and should specify why
the researcher feels access should be granted. For additional infor-
mation regarding appeal procedures, researchers should contact the
library holding the records.
(3) The appeal is first addressed to the Presidential library,

where the closed materials are re-reviewed by the library’s appeal
authority, often the director.
(4) If the appeal authority overturns the closure, the materials are
submitted for notification to the legal representatives of the former
and incumbent Presidents according to the Presidential Records
Act (PRA).
(5) If the appeal authority sustains the closure, the researcher may
appeal to the Assistant Archivist for Presidential Libraries. There
is no additional appeal authority.
Materials restricted due to national security can be challenged and are

reviewed by designated officials of the originating or responsible agency,
or by the National Archives and Records Administration (NARA) under a
written delegation of authority. Appeals to the agency responsible for the
declassification can be made and a researcher can further appeal a denial
under a mandatory review to the Information Security Appeals Panel.8
This process governs only those records that are national security–based
restrictions, not records that are unprocessed (which have to be requested
through FOIA).
After determining which documents are of interest, a researcher will be

asked to fill out a short form identifying the collection, box numbers, and
name of the individual in question (each of which are obtained from the
finding aid). The archivist will pull these boxes and bring the researcher
a cart containing the documents in individual archival boxes (most carts

can hold between fifteen and twenty boxes, and a researcher can usually
request only as many as the cart will hold). Because these documents
are often originals, they must be handled with care, but sometimes orig-
inal documents have been removed from circulation and photocopies are
included instead. Each archive will have a formal set of rules that must
be followed for a researcher to be allowed to view the documents. These
typically include having only one box out at a time, having only one folder
out at a time, having a folder marker in the box (to keep track of the folder
order in the box), and handling only one document at a time.
What a Researcher Can Bring

Most reading rooms will require that visitors leave backpacks, bags,
purses, and other personal items outside the reading room (typically in
library-provided lockers). As in many libraries, food and drink (and occa-
sionally gum) are generally not permitted. Most archives allow one to
bring in a laptop computer (although it may be searched when one leaves
the research room). Archives do not allow ink pens in the reading room,
but most will provide scratch paper (often with a hole punched in it so
that documents cannot be transported out) and a pencil for taking notes
and filling out any documentation. Special permission is usually required
for cameras or scanning equipment for making copies or duplicates of the
documents; this is becoming more common but is still reviewed on a case-
by-case basis. Ask the archivist or research room manager for details. Do
not be offended if the research or security staff asks to inspect a bag or
other materials. At one library, I had a security personnel ask me to open
my laptop to see if there were any documents inside. The job of these
staff is to keep the documents safe and available for current and future
researchers.
Photocopying
There are two ways to get photocopies from the documents in the
library: make them yourself or have the library make them. If a researcher
does the copying, the library usually has special procedures for making
these copies to protect the documents. Usually, a researcher will take a
whole box or folder (not an individual document) to the photocopy station,
which either has a code assigned to the researcher or a space for a counter
which is included to keep track of the copies. Only one photocopy may be
made at a time (so using the feeder attachment on the copy machine is not
allowed). These copies are (as of this writing) fifteen cents. If a researcher
prefers that the library make the copies, it is usually necessary to (1) fill out
a form specifying the collection, box, folder, and document to be copied
and (2) mark those documents with specially designated markers (marked
“go” and “stop” at some libraries). These copies are more expensive and
are (as of this writing) between fifty and seventy-five cents, depending on
the library. The copies will be mailed to the researcher, usually in the order
in which the library receives the requests. These requests can be made in
person at the library, over the phone, or by e-mail.
Archival�Challenges
The quality of archival research is only as good as the time the researcher
is willing to put in, the archivists, and the finding aids. A comprehen-
sive search of a presidential archive would entail sifting through several
hundred million documents. For comparative purposes, the Johnson
Library contains 35 million documents, the Nixon Presidential Materials
Project contains 45 million documents (only 7 million of which are open
to the public), the Ford Library contains 21 million documents, and the
George H. W. Bush Library contains 38 million documents. Many of the
documents in question are lost to history because of inadequate or incom-
plete finding aids. It is therefore necessary to rely rather heavily on the
archival finding aids, but thinking “outside the box” and pursuing alter-
native research strategies is often productive and necessary to find appro-
priate and germane data. For example, researching secondary sources for
alternative individuals or issues related to one’s research topic may be
fruitful. In one instance, I was having trouble locating polling data for a
primary presidential adviser who was sure to have access to public opinion
data. Secondary sources led me to the files of this staffer’s assistant, where
the archivists had filed the relevant documents.
A prominent disadvantage in using archival research is the possible

selection bias problem because certain papers may not be available for
research for a number of reasons. For all presidential administrations,
many small but significant events, such as personal thoughts or phone
calls, were not recorded (Howell and Pervenier 2001). Interpersonal
conversations or discussions in meetings often do not find their way into
the permanent historical record. Certain staff members may record their
observations from meetings but rarely are candid assessments included
systematically. This presents a challenge of compiling a complete record
of a decision-making time line, the causal factors that were taken into
consideration for a critical decision, or an analysis of the alternatives that
were considered in a particular situation.
Second, fundamentally, the records of the archives are those collected

by other people, often pursuant to statutes or rules that provide wide lati-
tude in what is released and what is not released. For instance, an archivist
at one presidential library may interpret issues in a particular memoranda
related to privacy differently than another archivist would. Simply put,
the nature of a specific kind of document released at one library may be
different at another library. Likewise, important and relevant documents
could have been lost or discarded either by the principal author or by
staffers when they were sorting materials (Hill 1993). Also, bulk mate-
rials (such as bulk letters to the president) are often destroyed because of
concerns about storing all of the excess documents. Archives keep less
than half of all papers that are related to the president, making the job of
an archival researcher challenging.
Third, access to certain papers may be restricted for a variety of other,

idiosyncratic reasons. Some of the materials that are uniquely relevant to
an analysis of presidential involvement with public opinion are not yet
open to the public. For instance, for later presidents, including George H.
W. Bush, Bill Clinton, and George W. Bush, many of the materials have
not yet been opened by archivists.9 During the first five years after a presi-
dent leaves office, the archival staff is tasked with the process of beginning
to process the files (usually around 5 percent of the collection). Pursuant
to the Presidential Records Act of 1978, after that five-year period is up,
those processed presidential records are made available for research and
all presidential records become open to FOIA requests.
Similarly, as discussed earlier, even if the material has been reviewed,

certain holds can be placed on the material based upon several reasons.
For instance, documents can be withheld because of a P-5 classification
(confidential advice from a staff member to the president) by an archivist
or a restrictive deed of gift. These possible reasons for restriction include
national security and privacy issues (papers relating to the appointment of
federal officials) as well as the fact that the release of the document may
violate a federal statute, disclose trade secrets or confidential commer-
cial or financial information, disclose confidential advice between the
president and his advisors or among such advisors, or invade personal
privacy. These determinations are made either by the archivists, jointly by
the archivists and national security staff, or by the donor of the specific
papers, who can put restrictions on the use of those documents. As noted
earlier, these papers can be requested to be released, but this process can be
slow. For older libraries, this is less of an issue. According to the National
Archives, “For libraries with papers and records over 25 years old, the
libraries have reviewed, declassified and opened a tremendous amount of
materials.”10 These reviews are conducted periodically and release more
than a million documents per year.
Interpreting�the�Archival�Record
Even if one has confidence in one’s search strategy, the sheer volume of
presidential documents precludes a full accounting of all of the documents
of interest. As a result, one can never be fully certain that a complete set
of documents can be found. For instance, in reading memoranda from

individuals inside the West Wing, one set of data may be interpreted in
one way by an individual adviser and in another way by a different staff
member. Similarly, specific key documents in the chain of evidence or
from important individuals may be missing, excluded, misfiled, or simply
destroyed. Relying on a single document for an evaluation of the White
House’s thought process can therefore be misleading. My experience in
researching the use of public opinion data by White House staff demon-
strated that some staff were much more likely to consult polling data and
to recommend to senior staff or the president that action be taken based
on that data. However, a more complete search for the important staffers
involved in advising the president demonstrates that the enthusiasm of
one staff member was often muted by the pessimism of another. For this
reason, using multiple perspectives or several case studies to gain perspec-
tive on the depth and scope of the discussion inside the White House
is important. In addition, it is wise to use supplementary primary and
secondary sources to corroborate archival findings.
Another issue is how the material that the president himself handled,
read, or commented on was categorized and filed. Documents which were
handled, read, or written on by the president are commonly stamped “The
President Had Seen” and retired to the president’s personal file. However,
specific presidents may have handled fewer documents and could have
been briefed more frequently in person than on paper. Presidential staff
secretaries or personal aides may have handled such documents differ-
ently in each administration because there is no rule governing how these
documents are identified as presidential property. The diversity of ways
in which these documents were handled makes causal inference problem-
atic. For example, I was interested in understanding the chain of evidence
concerning public opinion polling data within the White House. In partic-
ular, I wanted to know if the president personally inspected or read polling
reports as part of a way to establish that the president may have acted in
a certain way knowing public opinion. I ended up abandoning this as a
line of causal discussion because of the difficulty of systematically ascer-

taining whether or not each president was in contact with these documents.
In summary, the archival record on a subject, issue, or event may be
incomplete. The task for the researcher and analyst is to establish and
document a search strategy to demonstrate to readers that the conclusions
that were drawn were appropriate. There are several ways for a researcher
to achieve this. The preponderance of evidence all pointing in the same
direction can constitute evidence of a trend or causal mechanism (Yin
2002). That is, if the White House staff were giving the president consis-
tent advice to take a specific course of action and the president took that
action, the evidence should be considered strong. Similarly, an analytical
strategy that selects cases from a broad range of issues (e.g., economic,
budget, foreign policy) and from episodes involving different presidents
across time (e.g., mid-twentieth- and late twentieth-century presidents)
and across electoral circumstances (e.g., first- and second-term presidents)
allows for variation in the cases that are examined and thus a more persua-
sive generalizable trend.
Conclusion
A comprehensive presidential archival search is an informative way to
analyze presidential history and is central to many important works in the
study of the presidency (Stuckey 2006). A reliance on historical data for
the purpose of testing theories is a powerful way to examine trends in
the presidency. One major advantage of presidential archival research is
that empirical investigations can be conducted inside the “black box” of
presidential policy making. Further, as Page (1993) noted, untangling the
decision making of political actors may be best conducted by just such
a method because it provides a relatively clear and valid explanation of
the players’ thoughts and arguments. Armed with the fertile information
from the actual time of decision making, researchers can find out what the
president and his staff knew and when they knew it.
It is difficult, although not impossible, to treat archival data analysis

of this sort in a truly systematic manner. In that sense, as is the situation
with most case study–based works, it is impossible to definitively prove
the assertions that are made with qualitative evidence. However, given
the range of cases and the interpretations of theories of behavior and lead-
ership, it can be claimed with some certainty that assertions connected
to broad trends or to theories developed from the literature are plausibly
correct. The sheer volume of memoranda, letters, reports, and recom-
mendations prohibits a complete exploration, and one must be careful
when discussing casual connections and in proposing absolute conclu-
sions. Researchers and analysts of the trends must take care to set out
a firm and defensible research strategy before undertaking an archival
search. This will save time for a researcher and provide a justification for
his or her methodology. Even considering these serious issues, the rich-
ness, import, and uniqueness of the historical records that are housed at
these presidential libraries provide for an unparalleled analysis of presi-
dential behavior and executive activity.
Endnotes
1. http://www.loc.gov/rr/mss/guide/pres.html.
2. The Presidential Library Act of 1986 mandates that a library association
associated with each library provide an endowment to NARA for mainte-
nance (based upon the size of the facility). Congress thereafter appropri-
ates funding to the libraries.
3. http://bushlibrary.tamu.edu/research/finding_aids/whorm/.
4. http://www.fordlibrarymuseum.gov/library/guides/guidewhcf.asp.
5. http://www.jimmycarterlibrary.gov/library/pres_materials.phtml.
6. The Presidential Records Act (PRA) of 1978 makes defined presidential
records created after 1981 subject to Freedom of Information Act (FOIA)
requests.
7. This request is different from a FOIA request in that a FOIA request
officially asks the archive to process unprocessed documents related to a
FOIA request (e.g., Hilary Clinton and public opinion polling), where the
declassification process can be done on a case-by-case basis for papers
that have already been processed but have been restricted from public
access due to a sensitive security issue.
8. http://www.archives.gov/presidential-libraries/faqs/#18.
9. For instance, at the George H. W. Bush Library, some files in White House
Office of Records Management (WHORM) subject file were processed
systematically from 1993 to 1998. These are official government files
rather than papers donated to the library. Overall, most of the files are
not open. Many documents relating to the elections of 1988 and 1992 are
closed due to “personal privacy” issues.
10. http://www.archives.gov/presidential-libraries/faqs/.
Bibliography
Hill, Michael R. 1993. Archival Strategies and Techniques, Qualitative

Research Methods. Thousand Oaks, CA: Sage Publications.
Howell, Martha, and Walter Pervenier. 2001. From Reliable Sources: An
Introduction to Historical Methods. Ithaca, NY: Cornell University Press.
Page, Benjamin I. 1993. “Democratic Responsiveness? Untangling the Links
Between Public Opinion and Policy.” PS: Political Science & Politics 27:
25–29.
Rottinghaus, Brandon. 2008. “Presidential Leadership on Foreign Policy and
the Limits of ‘Crafted Talk.’” Political Communication 25: 138–157.
———. 2010. The Provisional Pulpit: Modern Presidential Leadership of
Public Opinion. College Station: Texas A&M University Press.
Stuckey, Mary. 2006. “Presidency Secrecy: Keeping Archives Open.”
Rhetoric & Public Affairs 9: 138–144.
Yin, Robert K. 2002. Case Study Research : Design and Methods. Thousand
Oaks, CA: Sage Publications.
Chapter 7
Rummaging
Through�the�Attics
of�the�CIA�and�Congress
How�Archival�Research�Enabled�Me�to�Write
a�Previously�Untold�Political�History
David M. Barrett
Getting�Hooked
More than a decade ago, while I was examining the papers of the late
senator Richard Russell (D-GA) in order to understand his influence
with President Lyndon Johnson on the Vietnam War, I learned that as
the chair of the Armed Services Committee and a senior member of the
Appropriations Committee, Russell knew more of the Central Intelligence
Agency’s (CIA) secrets than any other senator during the 1950s and 1960s.
Intrigued, I returned to the archive later on to explore the familiar ques-
tion: What did he know and when did he know it? Further, I wanted to
know if Russell had actually affected the CIA’s functioning or if he was

a paper tiger.
In many ways, my experience there foreshadowed what I would
discover at other archives with congressional papers—there were only
scattered documents relating to the CIA. Still, what was there fascinated
me: a transcript of a President John F. Kennedy–Russell conversation,
calendars showing Russell’s meetings with CIA leaders, the occasional
handwritten note from Russell to himself about something relating to the
CIA, and memoranda from staffers to Russell about the agency. Using
these, plus the Congressional Record and a few memoirs and histories,
I presented a paper on Russell and the CIA at an American Political
Science Association meeting and published an article (appropriately titled
“Glimpses of a Hidden History”) in an intelligence studies journal (Barrett
1998).
I was hooked, wanting to know, what had Congress done in rela-

tion to the Central Intelligence Agency in its early years? The secrecy
surrounding the topic made it all the more attractive to me, though I
would encounter tremendous frustrations over that governmentally regu-
lated secrecy. I also came to believe (thanks to the work of many scholars)
that congressional oversight of executive branch agencies sometimes
mattered (Ogul 1976). And, I eventually understood something basic: that
Congress has a legitimate role to play in relation to the functioning and
“dysfunctioning” of government bureaucracies. When Congress looked
into a failed U.S. Army battle with an Indian nation early in George
Washington’s presidency—thus pursuing an early case of congressional
oversight—it was exercising its constitutionally implied right to do so.
After all, Congress provided the money for military operations, just as it
does two centuries later, and it wanted to understand how well or poorly
that money had been spent.
By the end of the twentieth century, the very limited literature on

Congress and the CIA in the early Cold War era concluded that Capitol
Hill had been irresponsibly hands-off in neglecting the agency (Holt
Rummaging Through the Attics of the CIA and Congress 229
1995). Still, I had the impression from my research at the Russell Library
that on occasion, the senator had been inquisitive. His questions had some-
times shaped what CIA leaders (or a president, the agency’s boss) did.
My goal became clear: to write a book on congressional oversight of

CIA in the early Cold War years. Could such a book be written? I had
no idea. One thing I knew for sure, though: it could not be written if it
were based mainly on the available published records. It might be made
possible by unearthing records of CIA-Congress interactions in the papers
of the CIA and the late legislators who had held oversight responsibilities
toward the agency.
Diving�In
Despite the assertion in a few publications that Senator Russell had been
the congressional overseer of the CIA for a couple of decades, I doubted
that this could have been true. I assumed, more or less correctly, that
there were archives scattered around the nation holding papers of other
deceased legislators—heads of the Senate and House Armed Services and
Appropriations Committees—who had interacted with agency heads. I
was humbled, though, by the fact that although I had long known the
name of Richard B. Russell (who, for better or worse, was a Senate
giant on issues including national defense policies and civil rights), I
knew nothing about the other powerful legislators on the four committees.
Styles Bridges, John Taber, Clarence Cannon, and Millard Tydings?—I
had never heard of them.
I gratefully discovered the website “Biographical Directory of the

United States Congress, 1774–Present” (http://bioguide.congress.gov/
biosearch/biosearch.asp), which, among other things, guides researchers
to appropriate archives. I also found an equally valuable online bibliog-
raphy (http://intellit.muskingum.edu/), created by J. Ransom Clark, which
oriented me toward the substantial literature on intelligence topics. And I
read many good, if inevitably incomplete, histories of the Central Intelli-
gence Agency in its early period (e.g., Powers 1979; Ranelagh 1986). The
multivolume report of the famous Church Committee of the mid-1970s
was also a valuable (if sometimes overly critical) primer on what Congress
had done and failed to do regarding the CIA in the early Cold War era
(U. S. Senate 1976). Scholars of U.S. intelligence, especially Harry Howe
Ransom (1959) and Loch Johnson (1985)—who worked on the staff of the
Church Committee—provided me a much-needed education in the func-
tioning of and problems related to the CIA in the post-WWII decades.
Thus, there was a good and fairly substantial corpus of literature on the
CIA itself, and there were many excellent treatments of Congress in the
1940s through the 1960s by scholars and journalists (e.g., Fenno 1966;
White 1957). Still, there was precious little on Congress and the Central
Intelligence Agency. Beginning to think about research, I learned that the
National Archives in Washington, DC, has useful but fragmentary records
of committees which interacted with the agency, and National Archives II
in College Park, Maryland, has similarly limited but valuable CIA records.
Presidential libraries (in my case, primarily the Truman, Eisenhower, and
Kennedy Libraries) have reasonably good records relating to CIA and
congressional relations. But I thought that the make-or-break dimension
of my research would be what was available or nonexistent in archives
with the papers of once-powerful but mostly forgotten and now-deceased
legislators.
I quickly learned that some members of Congress had destroyed some

or all of their papers. Carl Vinson (D-GA), whose service on Capitol
Hill began during the Wilson presidency and who headed the House
Armed Services Committee in the early Cold War decades, retired in
1965. His committee had created a highly secretive subcommittee on the
CIA that became reasonably assertive in relation to the agency in the late
1950s. Upon retirement, he considered repeated offers from the Univer-
sity of Georgia to house his papers. Reportedly due to concerns about
those records revealing secrets that would endanger U.S. national security,
though, Vinson and an associate destroyed all of them (Cook 2004, 330–
331).
Until 1963, the office of Gerald Ford (R-MI), who served on the House
Appropriations Subcommittee on CIA, destroyed most of its records that
were more than a few years old. Ford’s reasoning was different from
Vinson’s: he simply wanted to rid his cramped office of “unnecessary”
papers. For reasons I could not discover, some committees had done the
same thing: at the National Archives, there are no papers of the House
Appropriations Committee from this era. I do not mean there are no papers
dealing with the CIA; I mean there are no papers of that era for the Appro-
priations Committee at all. Meanwhile, the House Armed Services’ papers
from the 1950s are so limited as to be almost worthless.
Also, as I began visiting archives, it became obvious to me that some

legislators, or their staff or heirs, had likely removed (or permitted the CIA
to remove) agency-related documents from their collections of hundreds
of boxes of documents. Two Appropriations chairs, Clarence Cannon (D-
MO) in the House and Carl Hayden (D-AZ) in the Senate, placed their
papers in archives, but those collections are of very limited value for
studying oversight of the CIA.1 In Hayden’s case, this may be mainly
because he monitored the agency so sporadically.
Clarence Cannon was another case entirely. His relationship with the
Central Intelligence Agency was different from what the conventional
wisdom had been about congressional oversight of intelligence in the
early Cold War period. I discovered that the irascible chairman summoned
Director of Central Intelligence (DCI) Allen W. Dulles to many a one-on-
one session in which Cannon raised critical questions. Similarly, his little
Appropriations Subcommittee on CIA kept Cannon’s attention across the
years, to the distress of agency personnel who were charged with congres-
sional relations. Though Dulles and other DCIs had a certain respect and
even affection for Cannon, people such as congressional liaison John
Warner (no relation to the subsequent U.S. senator of the same name)
frequently voiced exasperation over Cannon’s practice of summoning CIA
leaders to secret subcommittee hearings with less than twenty-four hours’

notice (Barrett 2005, 116–117, 257–259, 309–310, 316–318, 331–332,
375–376). However, I learned all of this not from Cannon’s papers, which
barely hinted that he played any role in relation to the agency, but from
the Central Intelligence Agency’s records at the National Archives.
In contrast, I have often sent my heartfelt thanks to certain late legis-

lators for not having their papers “sanitized.” For whatever reason, the
papers of Bridges (R-NH), Taber (R-NY), Russell, George Mahon (D-
TX), and some others have valuable and sometimes still “classified”
records of their interactions with CIA leaders.2
Early on, in my own university’s library, and before I understood

who Styles Bridges was—a sometime chair of the Senate Appropriations
Committee and a “walking 25 votes in the Senate,” according to an Eisen-
hower aide (Whitman Diary)—I came across a finding aid to his papers
in New Hampshire (Barrett 2005, 144). Examining it, I saw notations
of “Top Secret—CIA budget,” and so I had my graduate assistant (who
was conveniently heading home to New England) copy everything in that
folder and report back to me. She obtained letters and detailed documen-
tation of the CIA’s budget, especially for 1955—amounts, categories of
spending, places in the Defense Department’s budget where CIA accounts
were hidden, and so on. Common sense indicates that such information
would have been of no use in the early twenty-first century to Saddam
Hussein or Osama bin Laden, but the U.S. government absurdly kept
(and still keeps) most old agency budgets and innumerable other early
Cold War documents secret. This one file folder in Bridges’s papers was
immensely useful, allowing me to document that Appropriations leaders
did know and give direction to how the CIA spent its money. It permitted
me to give a fuller, more accurate description than had ever been given in
the literature of where CIA resources were being committed in the early
to mid-1950s.
Similarly, in Mahon’s and Taber’s papers, I found handwritten notes

of meetings with CIA leaders covering topics ranging from covert action
abroad to the agency’s views of an alleged “missile gap,” which Democ-

rats in the Eisenhower era spoke of, meaning that the United States seri-
ously lagged behind Soviet missile development.
Two�Uses�of�Documents
What can archives with papers of legislators provide to those researchers
with an interest in the functioning and oversight of intelligence agencies?
Two things are obvious: (1) the chance to engage in what Clifford Geertz
(1973) famously called the “thick description” of human cultures, and (2)
of great interest to many political scientists, the opportunity to count.
First, although my experience with archival research has often been

frustrating—due to the continuing government secrecy surrounding the
CIA’s activities many decades ago—those secrecy policies were often-
times not carried out with neglected papers of deceased legislators. There-
fore, I sometimes discovered intelligence-related memoranda between
legislators and their staffers. (Also, at National Archives II, I found many
CIA memoranda about the agency’s legislative liaisons’ interactions on
Capitol Hill.) This permitted me to add both color and nuance to my
descriptions, for example, of a Congress member’s anger at the CIA. It is
one thing to say that Senator Stuart Symington (D-MO), a big promoter of
the “missile gap” thesis, distrusted the agency for the way that it counted
Soviet missiles; it is another to be able to quote from a CIA staffer’s memo
on his encounter with Symington at a barbecue. The senator was furious
that Dulles had gone on television and said there was no gap between U.S.
and Soviet missile capabilities. Covert action chief Frank Wisner reported
to Dulles:
He was extremely annoyed with your having made certain state-

ments on this television program which you had…refused to make
to his committee…that you did not believe that the Russians were
ahead of us in the field of intercontinental ballistic missile devel-
opment…
Stuart then launched into a considerable review of the numerous

times and occasions on which he has come to the defense of the
Central Intelligence Agency. He reminded me of how he had stood
up against Senator McCarthy in our defense during and before
the McCarthy-Army hearings…and he recited other incidents, all
of which are, of course, accurate. He said that he had done these
things not merely because of his admiration and friendship for you
and certain others in the Agency, but because of his belief that the
Central Intelligence Agency was entitled to be treated in a special
and separate category. However, it if is the policy of this Agency to
engage itself on one side or the other of highly controversial polit-
ical issues, he would have to revise his views. (Barrett 2005, 246–
247)
Second, although I am not a particularly quantitative political scientist,

I certainly see the value in counting, especially in the area of legislative
studies. In my archival research, I learned that the CIA (created in 1947
by Congress and President Harry S. Truman) had not counted its interac-
tions with Congress until 1958, and then it did so imperfectly. But, by
drawing on its records and those of legislators, I was able to come up with
an approximate count of different types of encounters, year by year: CIA
leaders’ appearances at hearings, meetings between directors of central
intelligence and individual legislators, and so on.
In a sense, my book is a gigantic test of the hypothesis of Mathew

McCubbins and Thomas Schwartz (1984) that the amount and intensity
of congressional oversight of executive branch agencies in a given time
frame is primarily a function of the political environment. When interest
groups, citizens, journalists, or others “set off alarms,” Congress then steps
up its monitoring of the agency in question.
The counting allowed me to trace the ups and downs of Capitol Hill’s
oversight of the CIA from 1947 to 1961. I judged that such oversight
mostly fit the predictions of McCubbins and Schwartz. For example, in
1955, a relatively sleepy year with few alarms being set off, DCI Allen W.
Dulles testified about ten to twelve times. In 1958, American and interna-
tional politics were more highly charged following the late-1957 launch of
the Soviet’s first satellite, the 1958 riots that nearly killed then vice pres-
ident Richard Nixon and his wife Pat Nixon in Venezuela, and a coup that
overthrew a pro-American government of Iraq. These and other events
caused citizens and journalists (though not interest groups, because almost
none in the 1950s focused on Cold War policies) to prod legislators to ask
many questions about alleged failures at the CIA.
The DCI testified more than two dozen times in 1958, sometimes
against his will. He complained to his brother, Secretary of State John
Foster Dulles (who had a note-taker on another phone) that Senator
Russell—who usually was pro-CIA—“would not give him support not
to go” before the Senate Foreign Relations Committee; on the contrary,
Russell “thought AWD should go. AWD does not like to go, but does
not know what to do about it” (Barrett 2005, 297–298). Dulles also had
frequent private sessions with the few members who were authorized by
the Congress to monitor the CIA. Because of my time at the archives, I
could give a quantitative portrait of the rise and fall of congressional over-
sight activity across the 1947 to 1961 time frame while also telling some
good stories.
Another easily counted type of interaction is that occurring between

legislators and citizens by way of the mail. In the aftermath of the May
1960 Soviet shooting down of an American U-2 spy plane, House Appro-
priations chairman Cannon detailed and praised the spy flights over the
USSR, thereby becoming the first U.S. government official to describe
them publicly as a CIA program. His floor speech, given the day before
President Dwight D. Eisenhower first spoke to the nation about the event,
was widely praised in government and news media circles. Still, a count
of his office mail after the speech concretely reminded me that many
Americans were appalled by the secrecy and lies surrounding the intelli-
gence-gathering operation. Almost half of the twenty-six citizens writing
to Cannon berated him for, as one put it, “the same old unimaginative
display of jingoism” that had blighted so many recent statements from

U.S. and Soviet leaders (Barrett 2005, 398).
Advice
For those scholars considering doing research for the first time, especially
at an archive with the papers of a former member of Congress, I have a
few elementary pieces of advice. First, consult the previously mentioned
Biographical Dictionary website. Then,
1. Give careful thought to what sort of documentation is sought, and
then contact the archive holding the papers that are of interest.
Archivists and their assistants are usually warm and helpful. (This
may be less true at large archives with overworked staffs.) They will
know if a finding aid exists for the legislator’s papers. It may be
available online. Use this to create a list of boxes of folders that may
be the most fruitful for exploration. Give the archivists at least a few
days’ advance notice before arriving; some archives keep less-used
collections off-site and have to order their delivery to a researchers’
room. Unfortunately, I traveled to three well-regarded archives in
faraway states that somehow failed to have my requested papers
available for research on the day that I arrived, despite my advance
notice. Therefore, I recommend sending an additional reminder
about one’s arrival date and the collection and boxes one wishes to
examine on the day before one plans to get there. Also, inquire about
photocopying procedures and costs. Almost all archives now permit
researchers to use scanners and digital cameras to copy documents.
They are real time-savers at the archive, permitting one to maximize
the amount of materials that are examined there because the docu-
ments can be printed later on, if desired.
2. Make a written note of the archival location of every document
that is copied or that notes are taken from. This information will
be necessary if the document is cited. Also, keep a list of all of the
boxes that were looked at every day at each archive visited. If one
decides six months later that it is necessary to return to an archive,
it would be helpful to know what has already been inspected.
3. Do not be surprised to initially feel a bit overwhelmed at an archive:

one has just arrived, the collection of papers that is of interest is
350 boxes in size, and that finding aid was pretty vague. I can only
advise talking to the archivists, asking every imaginable question,
and diving in. It will become clear relatively quickly that some
boxes are irrelevant to the research topic, and one will move on.
4. Treat the documents with great care. Every decent archive’s rules
require this, and for good reason. I still have not gotten over my
shock at learning that former national security adviser Sandy Berger
mistreated, stole, and destroyed certain documents at the National
Archives (Leonnig 2005). Senator J. William Fulbright’s old book
title comes to mind: The Arrogance of Power. It is almost unimag-
inable to me that any serious scholar would do such a thing.
5. Do not be surprised if the archivist says, “The Congressman’s
papers may not have that information” or “Nobody ever comes
to look at Senator Jones’s papers.” This does not mean that his
or her papers will not be valuable. One should go and see for
oneself. Almost no one ever examines the papers of the late Senator
H. Alexander Smith (R-NJ), but when I did, I came across his
wonderful handwritten diary of life in Washington, DC.3
6. What should one do if one finds irregularities in the late Congress
member’s papers—for example, papers that are stamped “Top
Secret” with no other markings indicating that they have been
declassified? Due to my research topic, this has happened to me
occasionally. In those cases, I made my photocopies and said
nothing to the archive. Believe it or not, at a conference panel in
which I participated in Washington, DC, in summer 2011, govern-
ment officials made it clear that they still visit archives with
decades-old documents if they learn that such papers were never
officially declassified. Their first action upon arrival: pulling the
boxes with those papers out of circulation to researchers. Then there
is the risk that the process to determine if those papers should be
declassified will take not just months but years.
Great�Opportunities
The fact that other scholars rarely look at congressional papers means that
a researcher has the opportunity to do highly original research, analysis,
and writing. By finding documents in archives, I could demonstrate as
fact, not supposition, all sorts of things that were not known before
about the Congress and the CIA’s history. For example, after the agency
carried out extensive, costly, failed anti-Soviet covert operations in eastern
Europe in the early 1950s, CIA leaders told a congressional subcommittee
just that: the operations had failed (Barrett 2005, 96, 157–158). This coun-
ters the widely expressed claim in the literature that the CIA never talked
about failures on Capitol Hill and legislators never asked. Also, I learned
that subsequent to a pro-Soviet Czech government’s ease at putting down
an antigovernment protest without causing fatalities in the early 1950s,
DCI Allen Dulles privately expressed regret in 1956
that nobody got killed. I’d have felt much better about that, and
the Czechoslovakian people would have stood much higher in the
world’s estimation, if there had been a thousand or ten thousand
people killed in that. We kill more people on the roads every day
for no purpose. (quoted in Barrett 2005, 212)
Dulles’s cold-blooded remark certainly confirmed President Eisenhower’s

equally private remark about the CIA and Dulles: “Here is one of the most
peculiar types of operation…it probably takes a strange kind of genius to
run it” (quoted in Barrett 2005, 251). Finally, I discovered that before the
agency intervened in Cuba in the infamous Bay of Pigs episode, its leaders
ran those plans before at least two congressional subcommittees in the
spring of 1961. This and other documents showed that there was consid-
erable congressional pressure on Presidents Eisenhower and Kennedy to
“do something” about Fidel Castro’s Cuba (Barrett 2005, 440–445).4
It was obvious to me in going through the papers of the CIA at the

National Archives and the papers of late legislators in many different
archives that I was not accessing a systematic record of those parts of
the U.S. government; it seemed more akin to having gone through their
attics, where, mixed in with lots of junk and ordinary materials, I found
sometimes fascinating and important parts of their history that had been
forgotten, set aside, or nearly discarded. The papers I found have now been
donated to the Richard B. Russell Library at the University of Georgia
so that others may examine what I discovered.5 And, the book that I was
unsure could be written was published, all 540 pages of it, in 2005. It won
the D. B. Hardeman award for the best book of the year on Congress.
What distinguished the book, the prize selection committee said, was its
reliance on wide-ranging archival research.
Endnotes
1. The one value of their papers for my research topic is the constituent corre-
spondence. Cannon’s papers are at the Western Historical Manuscript
Collection, the University of Missouri, and Hayden’s are at the Hayden
Library, Arizona State University.
2. Bridges’s papers are at the New Hampshire State Archives, Taber’s are at
the Kroch Library at Cornell University, and Mahon’s are at the Southwest
Collection, Texas Tech University.
3. Smith’s papers are at the Mudd Library, Princeton University.
4. I reported one such subcommittee (of the House Armed Services
Committee) in the book; subsequent research shows that the CIA also
previewed its plans before the CIA subcommittee of the House Appropri-
ations Committee.
5. See the “David M. Barrett Research Files,” Richard B. Russell Library for
Political Research and Studies, University of Georgia. As of this writing,
I have retained documents relating to Cuba policies from 1959 to 1961,
which I expect to draw on for a volume devoted entirely to the John F.
Kennedy era.
Bibliography
Barrett, David M. 1998. “Glimpses of a Hidden History: Senator Richard

Russell, Congress, and Oversight of the CIA.” International Journal of
Intelligence and CounterIntelligence 11(3): 271–298.
———. 2005. The CIA and Congress: The Untold Story from Truman to
Kennedy. Lawrence: University Press of Kansas.
Cook, James F. 2004. Carl Vinson: Patriarch of the Armed Forces. Macon,
GA: Mercer University Press.
Fenno, Richard. 1966. The Power of the Purse: Appropriations Politics in
Congress. Boston: Little, Brown.
Geertz, Clifford. 1973. The Interpretation of Cultures: Selected Essays. New
York: Basic Books.
Holt, Pat. 1995. Secret Intelligence and Public Policy: A Dilemma of Democ-
racy. Washington, DC: Congressional Quarterly Press.
Johnson, Loch. 1985. A Season of Inquiry: The Senate Intelligence Investiga-
tion. Lexington: University Press of Kentucky.
Leonnig, Carol. 2005. “Berger is Likely to Face Fine.” Washington Post, April
2, A8.
McCubbins, Mathew, and Thomas Schwartz. 1984. “Congressional Oversight
Overlooked: Police Patrols Versus Fire Alarms.” American Journal of
Political Science 28(1): 165–179.
Ogul, Morris. 1976. Congress Oversees the Bureaucracy: Studies in Legisla-
tive Supervision. Pittsburgh: University of Pittsburgh Press.
Powers, Thomas. 1979. The Man Who Kept the Secrets: Richard Helms and
the CIA. New York: Knopf.
Ranelagh, John. 1986. The Agency: The Rise and Decline of the CIA. New
York: Simon and Schuster.
Ransom, Harry Howe. 1959. Central Intelligence and National Security.
Cambridge, MA: Harvard University Press.
U.S. Senate. 1976. Select Committee to Study Governmental Operations

with Respect to Intelligence Activities. Final Report. Washington, DC:
Government Printing Office.
White, William S. 1957. Citadel: The Story of the U. S. Senate. New York:
Harper.
Whitman, Ann. Diary. Dwight D. Eisenhower Library.
Chapter 8
Public�Opinion
in�the�Archives
Amy Fried
On my way to examine the papers of pollster Elmo Roper at the Univer-

sity of Connecticut, I was equipped with the simple paraphernalia of an
archival researcher and a relatively constrained research interest. It was
a cold day, wintry still, but with the strengthening sunshine that portends
spring. Following a trail from the legendary presidential election of 1948,1
when Dewey did not defeat Truman, I hoped these papers would help me
develop a clearer picture of a time of turmoil. The polling industry took
a blow from the 1948 election, yet it recovered and went on to flourish.
Perhaps Roper’s papers could help me understand how he and his compa-
triots countered a threat to the legitimacy of the polling enterprise.
In 1948, Roper, along with George Gallup and Archibald Crossley, was
one of the premier political pollsters of the day. “The “big three” poll-
sters each got their start not in the field of academic attitude research or
in the nineteenth-century journalistic straw poll tradition, but in the world
of business. They were first and foremost market researchers, devoted to
the science of improving corporate profitability through carefully crafted
adverting campaigns and public relations stratagems (Igo 2007, 113).
Roper was known to the public through his articles on issue polls in
Forbes and a Sunday evening CBS radio program. Like the other pollsters,
Roper had gained respect when he, in contrast to the Literary Digest, had
correctly predicted a win for Franklin D. Roosevelt in the 1936 election.
The Digest’s poor performance was a strong factor in the publication’s
demise and the fall of straw polls, as well as concomitant increased support
for so-called scientific polling2 using sampling methods.
Just twelve years after the Literary Digest failure, those scientific poll-
sters, having predicted a Truman loss for many months, were subject to
public derision. Although Gallup had touted polls as a mechanism of
democracy, the 1948 breakdown opened the door to critics of polling’s use
by political decision makers (Fried 2006; Rogers 1949). Some commercial
clients were restive and decreased their spending on polls (Converse 1987;
Moore 1992). Although all of the pollsters had given incorrect numbers
in advance of the election, Roper performed the worst, underestimating
Truman’s vote by twelve percentage points and overestimating Dewey’s
by seven (Mosteller et al. 1949, 17). How, then, did Roper and the nascent
polling industry survive this threat to the legitimacy of quantitative means
of discerning public opinion?
With folders from the Roper papers in front of me, I discovered corre-
spondence from members of the public as well as from academics, busi-
nessmen, members of the media, and an investigating committee set by
the Social Science Research Council. These papers helped me grasp the
size of the early polling community, the relationships among individuals in
different sectors, and the importance of preserving and forwarding polls,
as seen by varied members of the community. Visits to other archives have
Public Opinion in the Archives 245
filled out this picture. But my reading of documents did more than build
upon an initial interest.
Arguably more importantly, the raw materials of the past encouraged a
shift to a new theoretical lens. Though I came to the project concerned with
issues about legitimacy and its loss, the correspondence from varied actors
prompted me to think about the post-1948 period in terms of networks
of polling pioneers. Early pollsters from multiple sectors were entrepre-
neurial figures who built polling apparatuses and needed to gain and main-
tain concrete financial support to further their operations. To be sure,
social legitimacy was a resource to help them do so, but so were the series
of relationships among polling professors. Understanding the 1948 situa-
tion provoked a look backwards to see how these relationships were forged
and how polling operations were built by diverse organizations. Historical
institutionalism thus became the guiding theoretical perspective for this
larger research project, suitable because it emphasizes the role of the inter-
play among ideas, organizations, and institutions in producing changes;
the strengths and limits that are inherent in networks of individuals; and
the impact of initial decisions involving technologies, policies, and coali-
tional, institutional, or organizational arrangements on later developments
(see Ansell 2006; Lieberman 2002; Orren and Skowronek 2004; Pierson
2000a, 2000b; and Skowronek 1982). My adoption of historical institu-
tionalism grew out of documentary evidence in the Roper papers and
guided visits to other archives with collections related to polling and
survey research (largely) conducted in the United States during (roughly)
the 1930s through the early 1950s.3 Institutionalist literature focused the
analysis and pointed to scholarly approaches and literature for exam-
ining certain elements that were important to the development of opinion
studies, such as the enhancement of state capacity focused around citizens’
views and complex relationships among bureaucrats, policy makers, and
communities of color.
Those very same papers from the period have been used in the service
of different research questions linked to a largely cultural perspective.
Historian Sarah Igo’s self-described study of how surveys were a part of

“the making of a mass public” reviewed letters received by Elmo Roper
in the wake of the 1948 polling failure.4 Although many citizens were
critical of polling and skeptical of its representativeness, Igo contended
that “many individuals believed they counted in the eyes of the statisti-
cians” (2007, 160–161). For instance, one man was concerned that he was
partially at fault for what pollsters had so confidently reported, stating,
“I didn’t deliberately try to deceive. I changed my mind in the last week
of the campaign.” Although it was under challenge following this very
public fall, polling ultimately became a part of how Americans came to
define themselves as “average Americans,” their commonalities revealed
by the quantitative data that were collected and compiled by experts and
those they trained.
More broadly, this particular example suggests that archival research

on public opinion serves many overarching purposes. In fact, each
scholar’s approach can change over time, as mine did. This chapter next
discusses the uses of archives, points to specific collections and their use,
and considers particular advantages and disadvantages of this research
approach.
Archives�and�Research�in�Public�Opinion
One way of characterizing streams of social science research involves
comparing interpretive, qualitative research to positivist, quantitative
research.5 However, archival research can be used in diverse meta-
approaches to research and along with differing methodologies. Though
archival research is amenable to exploratory, inductive research, it can
also be associated with research focused on questions growing out of a
conceptual or theoretical scheme, including an approach emphasizing the
testing of hypotheses. Traditionally, archival analysis is qualitative, but
researchers can use quantitative methods. Furthermore, archival research
can be used to address many different questions about public opinion,
involving, among other things, the strategic uses of polls, the development
and use of quantitative means of understanding public opinion, and the
connection between policy development and polling.
Three public opinion studies, two focused on the presidency and another
on congressional leaders, display differing methodological approaches
involving archives. Contending that “polling stands at the heart of the
modern, candidate-centered presidential campaign, thus making it a fitting
proxy for evaluating the presence of a permanent campaign environment
in the White House,” Heith (2004, 73) investigated the use of polls in six
administrations, starting with President Richard M. Nixon. In conducting
this evaluation, Heith reviewed and coded documents preserved in collec-
tions of presidential papers. As she explained:
I use the memoranda to produce a distinctive and comparable

data set for each administration. Any memorandum, including any
White House memo that referenced poll data, contains a few basic
ingredients that characterize all memos. All memoranda include the
date the memo was sent, the sender of the memo, the receiver of
the memo, and the text, or information the memo contained. These
essential components reveal much about the employment of public
opinion data in the White House. I used these components to code
the memoranda from the Nixon, Ford, Carter, Reagan, and Bush
archives at the sentence level of analysis for the type of poll data
employed and the type of White House activities in which the data
was employed. (2004, 10–11)
By extracting information from memoranda, Heith was then able to

construct tables demonstrating similarities and differences across pres-
idencies. Most of Heith’s numerical presentations came in the form of
tables, charts, or graphs showing percentages (of, for example, which
administration offices exchanged polling memos and whether the memos
focus on the president’s popularity or several other phenomena). Further-
more, statistical analysis of the interrelationships among variables was
also presented. Applying findings from this analysis of polling memos,
Heith concluded that what appear to be campaign-like activities do not

interfere with traditional practices of governing.
Like Heith, Eisinger (2003) focused on presidents’ use of polls but
asked divergent questions and studied a different (although overlapping)
set of presidents. Beginning with President Hoover, “the last president
not to gauge citizens’ views via polls” (Eisinger 2003, 36), this study
placed presidents’ uses of polls in the context of interinstitutional rela-
tions. Congress and the president are often at odds, and at times, this was
reflected in views about whether polling was proper at all. For instance, in
February 1957, revelations that the State Department contracted to receive
polling data that were not shared with Congress provoked discussions
about institutions and representation (Eisinger 2003, 48–50). Besides
Congress, the media and political parties had historically made claims
about their roles in interpreting the public voice, but presidents did not
wish to be constrained by those interpretations. Although both Eisinger
and Heith worked with archival materials in analytical ways, Eisinger
did not translate qualitative information (i.e., documents) into categories
that could then be counted or manipulated statistically. Eisinger adopted
a traditional approach to archival documents as a means of under-
standing events, individuals, offices, and contextually based interactions.
Through these means, Eisinger drew conclusions about what adminis-
trations sought to accomplish in their quests for relatively autonomous
assessments of the public.
A third methodological approach is illustrated by Harris’s research

program concerning congressional uses of polls, which can be tracked
across a set of journal articles (see Harris 1998, 2005; Fried and Harris
2010). Harris has analyzed how the adoption of polls by congressional
leaders to shape policy agendas and their presentation was affected by
changing understandings of legislators’ democratic roles and shifts in
institutional contexts and strategic situations. With presidents and media
outlets able to gather their own polling data, congressional leaders began
to seek out their own sources of knowledge about national public opinion,
not the views of particular subnational constituencies. Evidence from

archival materials as well as interviews with congressional leaders and
staff members were used to support arguments about these dynamics.
For instance, in a discussion about how the monitoring of public opinion
informed agenda development, Harris (2005) drew from a memo from
Kirk O’Donnell to Speaker O’Neill stating, “If Peter Hart is right…then
the Democratic agenda from here on out is clear” (143), as well as from an
interview he conducted with a Democratic aide who pointed to polling as
a limit on legislative action. Buttressing documentary findings about the
place of polling within congressional leadership circles, a Democratic aide
from the 1970s and 1980s recalled, “If we had sobering polls…someone
would say, ‘We can’t ask our members to do this.…Sometimes that’s why
you’d pull it back [not schedule the legislation].’” This combination of
methods can be highly beneficial. Interviews add detail and texture to
what can be found from documents and help to establish narrative and
motives. To be sure, interviews are not always possible if one’s research is
placed further back in time or if participants will not consent to be queried.
Furthermore, using archival materials in addition to interviews prevents
too much of a reliance on individuals who have political concerns and
interests in mind when they speak about internal congressional matters.
Documents provide contemporaneous context to reconstructed memories
and personal accounts.
Although I have noted differing research questions explored by Heith,

Eisinger, and Harris, each fits into a broad orientation which I call the
historical turn in public opinion research. Although most scholarly work
on public opinion in political science involves the quantitative analysis
of poll data, various archival researchers have focused on political devel-
opments and change processes, encompassing research on (1) the place
of polling in governing, (2) the shape and impact of public commu-
nications for particular policy or political matters, and (3) the rise of
polling and survey research. Archival materials relevant to this approach
include a wide array of documents, such as summaries of particular
polls; public responses such as letters and phone logs summarizing calls;
memos regarding polls and poll use; planning documents and debates
regarding what to ask when conducting a poll; correspondence between
opinion researchers or between them and politicians, journalists, funders,
or people in the business world; and information such as personnel
records, proposals to carry out projects, organizational charts, bills, and
invoices, which show how polling operations were organized and funded.
Research on polling in governing has included work involving archives

(such as that of Heith, Eisinger, and Harris) and some which has not.6 One
important question has been the ways that elected officials and govern-
ment administrators respond to polling data and employ it strategically. An
account that placed this dynamic into a broader historical context argued:
Gathering data by way of polls served a variety of strategic

purposes for a burgeoning American state, such as designing
more effective policies, preventing resistance by the public to
federal intervention that affected local and traditional practices,
and shaping presidential rhetoric aimed at persuasion…[T]he later
adoption of polls by Congress was part of a shift toward plebisc-
itary politics that, like the rise of administrative politics, had its
roots in the Progressive movement. (Fried and Harris 2010, 323)
Furthermore, in a series of works, Jacobs and Shapiro (1994, 1995,

2000) have used archival research along with other methods to address
the strategic uses of public opinion in various policy debates as well as
the development of public opinion infrastructures. Opinion data are used
by presidents and legislators to determine how to frame arguments to best
promote their policy preferences and political ambitions.
In going to the archives, some public opinion researchers have looked

to communications from the public itself. Congressional and presidential
collections may include letters from constituents, and these have been
used to explore a number of important questions. Lee (2002) used letters
from the time of the civil rights struggle to address the claim that elites,
not citizens, are the predominant influence on public opinion.7 Lee coded
letters for demographic data and themes and used selections to present and
discuss what these letter writers cared about and thought. This approach
provides richness to the understanding of public opinion on civil rights,
effectively demonstrates the ability of citizens to find their own ways of
discerning political events, and enables a comparison of the ways different
groups of people approached and framed the issues at hand. In a similar
vein, historian David Thelen reviewed letters “from five thousand Amer-
icans to Congressman Lee Hamilton and the House Select Committee
on the Iran-Contra investigation that he chaired” (1996, 14). These citi-
zens’ views challenged the media discourse about events and about what
the public thought and demonstrated sophisticated visions of democ-
racy, heroism, and patriotism. Rottinghaus (2006) studied letters from
the public to various White Houses. Unlike Lee and Thelen, Rottinghaus
focused on the development of systems and apparatuses in the administra-
tions rather than the substance of citizens’ concerns. Using opinion mail
summaries compiled weekly by the administrations of Presidents Dwight
Eisenhower, John F. Kennedy, Lyndon Johnson, and Richard Nixon (and
which can be found in these presidential archives), Rottinghaus showed
an upward trend of mail and a migration of interest in citizen mail from
political entities outside of the presidency to within the White House.
A third strand of the historical turn in public opinion research has

focused on the rise of polling and survey research, in particular how and
why it occurred. Archival research by Igo (2007), as well as nonarchival
scholarly work by Herbst (1993), has emphasized the cultural and ideo-
logical elements of the shift to quantitative means of understanding public
opinion. Herbst has argued that polling’s adoption was enhanced by the
power of Weberian rationality, a construct which emphasizes scientific
means of grasping conditions of the world, as well as by the desire of
rational political actors to employ public opinion to further political strate-
gies and goals. With three case studies,8 Igo has pointed to the rise of
the idea of the “average American,” defined in large part by quantitative
opinion data. Jean Converse’s comprehensive study of the rise of survey
research (1987) relied on a mix of methods, including research in the
archives of several survey research centers, and provides an invaluable
look at the organizational and administrative efforts that were required

to give rise to and sustain these entities and the professional associations
and links which have provided professional support. Archival research in
this area has also demonstrated the personal and professional complexities
that are inherent in any nascent endeavor, whether that involved gaining a
home and financing for a new research center or working with sometimes
difficult colleagues.9
Another sort of interest in the history of public opinion could be called

forensic survey analysis. This approach uses polls and surveys from the
past, as well as contemporaneous analyses, to explore and explain past
public opinion as well as technical issues linked to the development of
quantitative opinion research. Work in the latter category includes studies
about the failed Literary Digest straw poll of 1936 which, as mentioned
earlier, launched the reputations of pollsters Gallup, Roper, and Crossley.
In 1936, Franklin D. Roosevelt won nearly 61 percent of the vote and
carried all states but two, although the Digest wrongly predicted that Alf
Landon would win with 57 percent to Roosevelt’s 43 percent. Ten million
ballots were sent out and more than 2 million people completed and sent
back postcards to the Literary Digest. In Chicago, about one-third of
all potential voters responded, as well as half of the registered voters in
Scranton, Pennsylvania (Squire 1988). A common explanation is that the
Literary Digest got it wrong because its sample was flawed, drawn largely
from phone books and lists of automobile owners. But the poll also failed
because it relied on people to decide for themselves whether to respond,
and the people who decided to send the ballot back were not representa-
tive of the population. As statistician Bryson pointed out, “even in the
depths of the depression, phones were not all that unusual” (1976, 184),
and the size of the error cannot be explained by this factor. With only one-
quarter of the people who received the ballot responding—or what would
be considered a low response rate today—those who displayed the moti-
vation to reply may have been those who were most dismayed with the
incumbent. And, even if these problems had not existed, the Digest poll
kept a running total of responses by area, adding new numbers to old ones
rather than reporting the results for discrete periods of time. By aggre-
gating numbers, the Digest neglected the possibility that opinions might
shift during the course of the campaign.
Responding to Squire’s (1988) analysis of problems with the straw poll,

Cahalan (1989) brought forward relevant data from his 1938 MA thesis.
In addition to having used more easily available national data from a 1937
Gallup survey, Cahalan “also conducted a probability sample telephone
survey of persons on the Digest’s mailing list in a medium-sized city in
Iowa within three months after the 1936 election” (1989, 129). Cahalan
noted:
Ironically, the thesis has reposed for these 50 years in the Univer-
sity of Iowa Psychology Department library, located within easy
walking distance from Squire’s own office. The findings were also
summarized in a Psychological Record article the following year…
However, Squire’s not being aware of this research is understand-
able because I don’t think my findings were ever cited elsewhere.
At that time [Public Opinion Quarterly] was just getting started and
neither [the American Association for Public Opinion Research]
or other mechanisms for dissemination of information on survey
methodology yet existed. Also, the handful of pioneers engaged
in research in those depression days were too concerned about
keeping their own organizations alive to afford the luxury of the
further flogging of what they considered to be a rather dead horse.
(129)
Beyond the particularities with the Literary Digest’s straw poll, it seems
certain that there have been myriad state and local polls conducted in the
past which may have much to offer researchers but which are not gener-
ally accessible and are tucked away in archives and libraries. Although
national surveys and polls can be found more easily, their analysis is
not straightforward by any means. Although they avoided the difficulties
inherent in straw polls, opinion researchers in the 1930s and 1940s used
problematic methods of sampling. Today’s opinion researchers use proba-
bility samples, in contrast to the quota sampling of the past, with discretion
left to fieldworkers about how to select individuals to fill demographic

quotas. Because the choices were open to the interviewers, they were able
to decide to question people in houses that were relatively well-kept and
individuals who seemed more accessible and approachable, thus intro-
ducing a decidedly nonrandom element to the sample. With each inter-
viewer making individual decisions of this sort, biases related to such
choices could be embedded in the aggregate data. Furthermore, Gallup
sought samples that would be representative of voters, not citizens as a
whole.
Women, southerners, and blacks voted at low rates in the 1930s and
1940s. These groups were therefore deliberately underrepresented
in the Gallup samples. For instance, from the mid-1930s through
the mid-1940s, Gallup designed his samples to be 65 to 70 percent
male. (Berinsky 2006, 506)10
Although using data from this period would be helpful to understand shifts
in the mass public, the data need to undergo statistical transformations to
become more valid and reliable.11
Collections�for�Archival�Research�on�Public�Opinion
Given the range of questions and possible foci involving public opinion,
there is consequently a great diversity of potential archival sources. Those
interested in issues growing out of the historical turn in public opinion
research might use multiple collections, which could include the papers
of presidents, federal agencies, members of Congress, organizations, and
foundations. Elsewhere in this volume (chapter 5), Rottinghaus discusses
the location, organization, and use of presidential papers. As I have already
discussed, political scientists have given these attention in their public
opinion research. Still, there is always more that can be done with these
collections, both because they are so extensive and because a researcher
may focus on a specific policy issue, a period in one or more presidencies
(such as the first year in office), or the relationships between a presidential

administration and the media or specific agencies, groups, or individuals.
In researching public opinion issues related to one or more presidents,
researchers can go beyond the papers in presidential collections. Presi-
dential collections include papers from some people associated with the
administration, but others’ papers are in archives elsewhere. Though the
researcher can use a good Internet search engine to locate paper collec-
tions of people he or she has identified as important to a president’s efforts,
archivists may have already provided guides for scholars. For instance,
the Franklin D. Roosevelt Library provides a guide to “FDR collections
at other repositories,” which lists 173 people and the appropriate web-
based sources of information about particular archival holdings.12 A good
many of these paper collections can be found in Washington, DC, at
the Library of Congress; however, others are scattered across the United
States at various universities and libraries. And, although most individ-
uals’ papers are in one location (or at least one city), others are divided
between several places. For example, the papers of Bernard Baruch, a
member of Roosevelt’s Brain Trust, are available at Princeton University
and the Library of Congress.
Because members of the U.S. House of Representatives and Senate

have often taken an interest in public opinion and its measurement by
polls, archival holdings related to individual members of Congress can
be used to further certain research projects. Papers of individual members
of Congress tend to be located in their home states, and finding them
can take a certain degree of effort. The National Archives includes the
Center for Legislative Archives, which maintains a list of congressional
collections around the nation.13 The center and other archives recommend
two print sources on such holdings, Guide to Research Collections of
Former Members of the United States House of Representatives, 1789–
1987 and Guide to Research Collections of Former United States Sena-
tors, 1789–1995. Information about U.S. legislators can be found in the
Biographical Directory of the United States Congress. The Senate Histor-
ical Office, which should be consulted when researching a senator and

historical aspects of the Senate,14 notes that
the origins of the Biographical Directory go back to 1859, when

Charles Lanman, an author and former secretary to Daniel Webster,
assembled the first collection of biographies of former and sitting
Members for his Dictionary of Congress. With the creation of the
Senate Historical Office in 1975 and the House Office of the Bicen-
tennial in 1983, professional historians assumed responsibility for
revising and updating the Directory. The results of their efforts
appeared in the bicentennial edition of the Biographical Directory,
which appeared in 1989. Since November 1998, the print edition of
the Biographical Directory has been complemented and enhanced
by an online database that is frequently updated.15
Political Graveyard (http://politicalgraveyard.com/), which was devel-

oped and is maintained by a private individual, Lawrence Kestenbaum,
the county clerk and registrar of deeds for Washtenaw County, Michigan,
is an extremely useful guide to information about federal, state, and local
governmental officials.
It was through some of these sources that I was able to find the loca-
tion of the papers of Representative Walter M. Pierce (1861–1954), a
former governor of Oregon who served in the House of Representatives
from 1933 through 1942.16 Pierce was of interest to me because he was
a harsh critic of straw polling and the polls which replaced them and he
sought to regulate and limit them. A variety of publications in the 1930s,
ranging from newspapers to Public Opinion Quarterly, took account of
his criticisms. Pierce’s papers are housed in the Special Collections of
the University of Oregon Libraries in Eugene, which has an excellent
finding aid online. Using this delineation of the collection, I identified
relevant files and ordered copies of the documents in them. Being able
to have documents copied by the archivist and mailed to me, without
me having to visit the collection, saved considerable money. Though the
copies were more expensive than they would have been if I were at the
archive, there were no travel costs. Of course, it is possible that I missed
some papers that would have been helpful, but my confidence in being
able to gather essential items was supported by the archive’s clear identi-
fication of files on public opinion as a topic area as well as on the legis-
lation to restrict polls that Pierce offered. It was buttressed by the variety
of materials that were forwarded to me, ranging from clippings to notes
on meetings with administrative officials and other legislators to corre-
spondence from citizens, scholars, and pollsters. This investigation also
benefited from archival materials that were available though the Center
for Legislative Archives. In February 1935, Pierce introduced H.R. 5728,
a bill that would bar straw polls from the U.S. mails, whether in a “letter,
writing, circular, postal card, picture, print, engraving, photograph, news-
paper, pamphlet, book, or other publication, matter, or thing.” Through
the Center for Legislative Archives,17 I obtained (again, upon request to
the archivist and with receipt by mail) relevant correspondence between
the chair of the Committee on the Post Office and Post Roads and James
A. Farley, the postmaster general. This example points to the availability
and means of accessing archival materials on congressional matters and
public opinion, even for a relatively unknown House member. Legislative
leaders’ papers are more extensive, and their interpretation is assisted by a
larger amount of related historical materials, including oral histories about
their work in leadership and on particular policies.
In investigating the uses of public opinion data in policy development

and implementation, scholars may wish to turn to archival materials from
federal agencies. The National Archives is responsible for maintaining
these collections, and it provides the finding aids for many of them online.
Relatively few papers have been digitized, and researchers must often
visit the appropriate center to read, copy, and take notes on the materials
they need. It is also possible to order specific papers and receive these by
mail or to hire a researcher to work through the files and provide docu-
ments. The National Archives maintains a list of private researchers by
geographic location, subject area, and media type.18
Because survey research has been supported by foundations and

because people who conducted and oversaw opinion research have
founded and worked with various organizations, some scholars may wish
to investigate the archives maintained by these kinds of groups. As I have
been interested in the relationships between scholars Paul Lazarsfeld and
Hadley Cantril and their work on radio audiences, and some of this work
received foundation support, I visited the Rockefeller Archive Center
(RAC) in Sleepy Hollow, New York. The RAC has relevant papers from a
number of organizations, including the Social Science Research Council,
the Russell Sage Foundation, and the Rockefeller Foundation.19 Archival
materials generated by a major political party are sometimes housed with
presidential papers from that period. Other organizations’ papers, if they
are archived, can also be useful for scholarly research on the development
and use of public opinion.
A number of archives maintain oral history collections, which can be

extremely useful in understanding how public opinion data have been
gathered and used. In studying the development of polling and survey
research, Columbia University’s oral histories are an excellent resource
because they include interviews with, among others, pollster George
Gallup, media researcher (and later president of CBS) Frank Stanton,
Social Science Research Council president Pendleton Herring, and scholar
Paul Lazarsfeld. Furthermore, multiple oral histories were conducted as
a part of broader projects on particular endeavors and individuals, such
as the Hollywood film industry, radio pioneers, black journalists, Social
Security and Medicare, and the Eisenhower administration. Although
many of these oral histories can be read only in the Oral History Research
Office in the Butler Library on the Columbia University campus, others
can be ordered on microfiche and some are available online.
Governmental officials’ collections often include oral histories from

people who were involved in their work, including those with an interest
in public opinion. Because of the ability to remotely search and access
them, those in the Truman Library are highly accessible and usable. For
instance, knowing that scholar Hadley Cantril had been a source of poll
data in Franklin D. Roosevelt’s administration, I searched for his name at
the Truman Library’s website and found several oral histories, including
one with John S. Dickey, who held various positions in the Roosevelt
and Truman State Departments.20 His interviewer, Richard McKinzie,
explored the use of polls prior to the 1945 United Nations Charter confer-
ence in San Francisco. These were “opinion polls which the Department
had made through Hadley Cantril and some other people,” and McKinzie
asked, “It’s a matter of judgment, of course, but how seriously were
opinion polls taken then and by whom?” Dickey replied:
Well, this is a fair question, because they were not taken seriously,
in my opinion, widely in the Department. The value of polling
at that point was still somewhat arguable, much more so than it
is today. Today, it’s accepted as a useful tool of politics and the
social sciences—if well done, it is valuable, even essential. And
we were carrying it on undercover. We were using, of course, all
the public polls of Gallup and so forth, but we also were using the
confidential funds of the Secretary’s office to compensate Lloyd
Free and Hadley Cantril, two academic polling pros, and Jerry
[Jerome S.] Bruner, to work with us. Bruner subsequently became
widely regarded as a psychologist at Harvard. Those are the three
who worked with us, formulating questions, etc.…These were
very sophisticated, competent people. Particularly, we were using
the polls to find out the level of information of American public
opinion during the war and on post-war planning and the areas of
ignorance, in order to help us develop more effective public infor-
mation programs. We wanted to know what needed attention in a
speech by the Secretary or others, or in Department publications.
In our polls we were running some basic, continuing questions.
For example, do you believe that we can trust the Russians in the
future and in postwar organization planning? We would compare
the findings, asking the same question about every three months,
keying the questions to events and comparatively to other nations.
I think the polls were taken very seriously by those of us who were
working on them in the public information work—in other words,

to guide our public information activities.
As Dickey’s interview indicates, public opinion polls were used to guide

not policy but rather its presentation. Relying on the work of academics
to craft the wording of poll questions and potential responses, polls were
a tool for understanding and trying to understand public views and their
dynamics and went into the mix along with weekly reports on editorials
and news reports. But, said Dickey,
I don’t think they were taken too seriously by the Secretary, the
Foreign Service officers, and others. If they happened to coincide
with what they thought was public opinion, oh, then they would
cite them to the Congressmen and to others.
Dickey’s comments are consistent with how a number of scholars in

the past few decades have seen polling used. Further research with oral
histories and archival collections can define the various ways polls were
used, by whom, and with what policy and political effects.
Political scientists engaging in forensic survey analysis must look to

various collections of survey data. The Roper Center’s Public Opinion
Archives has the greatest variety of polls over the longest period of time.
For example, it houses nearly 150 polls with data collected between
1935 and 1940. Full access to data sets is available to member institu-
tions, and scholars elsewhere can purchase specific data sets. The Amer-
ican National Election Studies and the National Opinion Research Center
house other important collections of survey data. Researchers should
contact these organizations for information about obtaining these data.
Using�Archives
Archival research often feels like being a part of a treasure hunt. Even
with a plan, one never knows what one will find, and there are periods of
elation and drudgery, yet it is essential to be prepared and stay focused.

Based on my experiences, I can offer these tips.
First, arrive at the archive as prepared as possible. At a minimum, know
whether the collection to be investigated is on-site or will have to be
retrieved from another location. If the latter, the archivist will probably
need at least several days’ notice of one’s visit. Some archives require
that a researcher inform them of a visit and reserve space ahead of time.
Even if a reservation is not required, I have made it a practice to e-mail the
archivists to introduce myself and my research project. Contacting them
generates a contact at the archive whom I can call upon when I am there
and after my visit, and it gives me a sense of comfort that they and I are
prepared for my visit and my time and the expenses of the trip will be well
spent. In many cases, there are excellent finding aids online which can be
used to prepare a preliminary list of the boxes and files one will want to
search. Of course, being physically at the archive may reveal other mate-
rials for examination. Another element of preparation is to have done as
much background reading as possible in order to develop an initial under-
standing of the broader context of particular documents as well as the role
and significance of specific individuals or organizations.
Second, make sure to develop a system for keeping track of findings

and specific documents. Each archive has its own methods and rules for
reading and copying documents. These can be extremely picky, such that
in one archive slips of paper inserted before a document to be copied must
be placed so that they stick out of the top of a file, whereas in another place
they must be placed sideways. Although these practices are arbitrary, it is
imperative to use each archive’s system, both to preserve a good working
relationship with their staff and to make sure that one will be able to get the
copies one wants. Some places will allow researchers to copy some docu-
ments (and this can vary depending on their length), whereas others do
all of the copying themselves. Further, researchers may be able to photo-
graph or scan documents, thus saving the cost of copying as well as the
time of filling out copy order paperwork. If photographing or scanning
documents is a part of the plan, make sure this is an acceptable practice

beforehand and save the pictures as soon as possible to different media as
well as to a web-based photo service.
Third, when looking for something specific, keep in mind that creativity
and persistence are necessary—and even then the item may not be located.
At one point, I was looking for a report that was referenced by an early
opinion research leader and an individual who worked for him, which
both used as an example of why the U.S. government should study public
views.21 Many of the local people were biracial, both African American
and Native American, or what whites derogatorily called “redbones.”
Fires were being set in Kisatchie National Park in Louisiana and offi-
cials wanted to stop them. Although I looked assiduously in the National
Archives in College Park, Maryland, the finding aids for the appropriate
agencies were not detailed, and despite pulling many boxes and looking
in many files, I could not locate the study. But I did not want to give
up the hunt. Back in my home office, I went to the Kisatchie National
Forest’s website, found contact information for the Forest Heritage and
Tribal Program manager, and wrote, asking if she could possibly help me.
I received not only a copy of the report but also a photocopy of an out-of-
print book about the history of Kisatchie.
Fourth, after returning home, one should make sure to update one’s
filing or notation system. When working on the book Pathways to Polling,
I created files for documents and notes organized around each planned
chapter. The system also includes a master document that states which
documents (or groups of documents) are in each file. These allow for their
(relatively) easy retrieval for the writing process. Make sure to take time
after each visit or after receiving documents by mail to incorporate them
into the working files and system.
Fifth, one’s approach should combine a certain obsessiveness about

details with an openness to new discoveries. Let the obsessive self loose
when it comes to keeping track of documents. Be sure to find out the
archives’ preferred citation method and make sure this information has
been included with whatever system is being used to take notes and track
copied (or photographed) documents. The same applies for documents
that notes have been taken on but that have not been copied. In addition,
keep a file on whether permission will be required to quote from or cite
documents from each archive and collection. To make sure I am gaining
as much as I can from my investigations, I regularly consider what these
documents imply about the broader research questions I am exploring.
Depending on what I am finding, my initial focus and working hypotheses
may need to be modified or shifted. (Other researchers might not find this
useful, particularly if they are coding all documents across archives and
collections according to common categories.) At various intervals, I have
made it a practice to jot down some notes about what I have found. This
enables me to capture insights and to identify particular documents which
elucidate particular dynamics.
Finally, one should make sure to count historians among one’s friends,
or at least close colleagues. They have considerable experience in working
with archival materials and can be wonderful sources of support and
concrete assistance.
Archival�Challenges�and�Possibilities
Comprehending the past requires documents generated by others for
different purposes and using different tools than the scholars who study
them. For scholars whose work is part of the historical turn in public
opinion research, archival investigations are essential for understanding
and explaining internal decision making, intraorganizational processes,
and interactions among institutions, organizations, and political actors.
Researchers who are interested in forensic survey research engage in
sophisticated methodological transformations of quantitative data. What-
ever the approach, recovering and interpreting the past is always fraught
with difficulties. As historians and others note, archives preserve partic-
ular documents that reflect certain vantage points. Furthermore, decisions
about what is kept and how documents are organized reflect constructed
structures of knowledge and power. Although there is not room in this
chapter to explore these meta-issues, it is worth noting a set of discussions
regarding the connection between what is found in archives and the ability
to develop rational, well-grounded arguments about the past.22
These larger concerns should provoke archival researchers to be self-

aware of the challenges and possibilities related to their sources and
methods.23 Take, for example, documents about polling that can be
found in many presidential libraries’ collections. Without a doubt, polling
memos and reports are a treasure trove of information, illuminating the
development of presidential messages and policies. These documents
were not created for public consumption, and administration officials,
given the desire to appear untethered from polling, did not want their
internal strategizing to be publicized. However, their apparent status as
promoting the unified view of the administration is open to question.
Scholars could do more in discerning political conflicts within the admin-
istration regarding priorities and approaches. Presidential administrations
contain internal schisms and alliances, and internal discussions of polls
may reflect these divisions. Political uses of polls are affected by preex-
isting policy commitments, themselves rooted in political interests, coali-
tions of supporters, and the desires to signal to key public constituencies
and to “sell” particular policies.24 To be sure, the actual mix of documents
contained in a particular presidential archive may or may not lend itself to
this sort of analysis. But, in any case, exploring these dynamics requires
an examination of documents while keeping in mind a mix of internal and
external political dynamics.
Further, no matter what analytical scheme is used in investigating docu-

mentary materials, archival research is but a part of a well-developed and
theoretically oriented investigation. As with any scholarly effort, the mate-
rials that are considered by archival researchers can change over time and
can come to include additional data, even as the analysis should be focused
and rigorous. Researchers must establish relevant contexts for interpreting
what they find in the boxes of files, including such items as economic
data, congressional hearings, election results, and information about social
groups. Political scientists trying to understand the place of public opinion
in politics who turn to archives to do so face a complex but worthwhile
task.
Endnotes
1. See Karabell 2000 for an excellent account of the 1948 presidential elec-
tion, the political and coalitional complexities which gave rise to four
significant candidates, and its aftermath.
2. This term was used by political scientist and survey research pioneer
Hadley Cantril. See Hadley Cantril, “The Bombardment of Ballots,” The
New York Times, June 14, 1926, SM6, and Hadley Cantril, “Straw Votes
This Year Show Contrary Winds,” The New York Times, October 25, 1936,
E3.
3. These have included visits to the Rockefeller Archive Center for papers
of the Rockefeller Foundation and the Social Science Research Council,
the Franklin D. Roosevelt Library for papers from multiple collections
involving polling, the University of Chicago Library Special Collections
Research Center (to read papers by, among others, Ernest Burgess and
Philip Hauser), the National Archives in College Park, Maryland (to read
papers from the Department of Agriculture and the Office of War Infor-
mation), and Columbia University’s Rare Books and Manuscript Library
(to read a variety of oral histories and collected papers, including those
of scholars Paul Lazarsfeld and Lindsay Rogers). I have also used mate-
rials from the Truman Library, the Center for Legislative Archives, and
the collected papers of Congressman Walter Pierce of Oregon from the
Special Collections & University Archives of the University of Oregon
Libraries.
4. See Igo 2007, chapter 4.
5. Political scientists who use qualitative methods frequently do so in the
service of interpretive social science, which focuses on subjective under-
standings. Although interpretivist researchers begin with research ques-
tions and categories, these are open to modification as research proceeds.
Sociologists Glaser and Straus (1967) argued on behalf of “grounded
theory,” which arises from and is grounded in research discoveries. Partic-
ipant observation, intensive interviews, focus groups, and Q-sorts are
methods commonly associated with this approach in political science. In
contrast, since the rise of the behavioral revolution in political science
(Somit and Tanenhaus 1967), many scholars of politics have oriented
themselves around a template borrowed from the physical sciences.
Researchers in this tradition develop hypotheses rooted in a conceptual
or theoretical framework, craft indicators related to concepts and vari-

ables related to the broader framework, and then test those hypotheses.
This positivist model operates with deductive reasoning and can provide
powerful explanations backed by statistical analyses, but it can fall prey to
problems related to using data that are not consonant with key concepts,
a lack of attention to subjects’ points of view, and difficulties in capturing
historical contexts and processes.
6. For instance, I have argued that perceived public opinion is a political
resource and examined legislators’ responsiveness to constructed public
opinion in various situations, particularly the Iran-contra hearings of 1988
(see Fried 1997).
7. The analysis is based on a “sample of 6,765 letters written to the president
concerning civil rights and racial equality between 1948 and 1965” (Lee
2002, 14).
8. These include the impact of and public response to the Middletown studies
by Robert and Helen Lynd; the pollsters, particularly after the 1948 pres-
idential race; and Alfred Kinsey’s studies of human sexuality.
9. See Converse 1987. Paul Lazarsfeld, an Austrian expatriate, was an
important figure in early survey research who appeared to have had
some problems with collegiality, some due to his temperament and some
generated because of cultural differences and a certain anti-Semitism.
See Stanton 1991; Lazarsfeld 1982, 37–38; and Letter, April 27, 1934,
from John V. Van Sickle, program officer, Rockefeller Foundation to
Robert Lynd, Department of Sociology, Columbia University, in the Paul
Felix Lazarsfeld Papers; series 1, box 27, folder Rockefeller Founda-
tion Fellowship 1, Rare Book and Manuscript Library, Columbia Univer-
sity Library. Van Sickle wrote to thank Lynd for his letter regarding a
renewal of Lazarsfeld’s fellowship and noted, “Your judgment of Lazars-
feld seemed to me excellent save on the one point of his appearance. It is
some years since I have seen him but my recollection is that he definitely
bears the mark of his race.”
10. See also Robinson 1999 and Igo 2007 on the consumerist view implied
by Gallup’s polling.
11. Berinsky noted that data from the 1930s and 1940s “are extremely difficult
to work with. Most of these surveys have not been touched for almost 60
years, and as a result, the data sets are not easily usable” (2006, 500). He
provided a weighting method to correct problems with these data. See also
Rubin 1956 on the possibility of reweighting data taken from straw polls,
another example of a “defective sample.”
12. For more, see http://www.fdrlibrary.marist.edu/archives/resources/

repositories.html.
13. See http://www.archives.gov/legislative/repository-collections/.
14. For more, see http://senate.gov/artandhistory/history/common/generic/
Senate_Historical_Office.htm. In addition to providing information about
particular senators and their collected papers, the office has developed oral
histories and historical essays, statistics about the Senate, and collections
of photographs and art. The historian of the House of Representatives has
some, but fewer, materials online. See http://historian.house.gov/.
15. Via http://bioguide.congress.gov/biosearch/biosearch.asp, one can locate
information about House members and senators and see lists of research
collections.
16. These sources also pointed me to a very useful doctoral dissertation about
Representative Pierce as well as several other writings about him.
17. Regarding the records of the House of Representatives, the Center
for Legislative Archives notes, “Before their transfer to the National
Archives most records of Congress had been housed in the offices,
attics, basements, and storage rooms of the Capitol. They had suffered
from neglect, vermin, and pilferage, abuses common to most collec-
tions of older Government records housed in unsuitable and unsu-
pervised storage areas.” At this time, records are considerably safer.
Although some have been digitized and therefore can be accessed
relatively easily, adequate funds have not been allocated to complete
this process. However, I have found the staff to be exceedingly
helpful and worth consulting. See also http://www.archives.gov/legisla-
tive/guide/house/chapter-01-research-strategies.html regarding strategies
for researching congressional records and collections.
18. See http://www.archives.gov/research/hire-help/.
19. As an aside, the building has high-quality art on the walls downstairs, as
well as colloquia with researchers and a nicely equipped kitchen where
one can take one’s lunch and make tea. Scholars peruse these materials
on the second floor of the facility. Down a small hill from the archive is a
church with Chagall windows purchased by the Rockefeller family. Space
in the archive is limited and must be reserved beforehand. The Rockefeller
Archive Center has an annual grant competition for grants up to $5,000.
20. The interview was conducted in Boston on July 19, 1974, and is available
at http://www.trumanlibrary.org/oralhist/dickeyjs.htm.
21. Likert 1948; Skott 1943. Also see my discussion of the study and the inci-
dent that provoked it (Fried 2012, chapter 3).
22. See Appleby, Hunt, and Jacob 1994; Burton 2005; Foucault 1982.
23. Challenges relating to forensic survey research include methodological
issues about how to correct data that today are considered to be of low
quality because of the sampling approaches that were used. In addition,
as with all polls, there may be issues with the construction of the survey
instrument, particularly question wording. Finally, there are many state
and local polls that have been largely overlooked but deserve attention.
24. See Fried and Harris 2010; Jacobs and Shapiro 2000.
Bibliography
Ansell, Christopher. 2006. “Network Institutionalism.” In The Oxford Hand-

book of Political Institutions, edited by R. A. W. Rhodes, Sarah A. Binder,
and Bert A. Rockman, 75–89. New York: Oxford University Press.
Appleby, Joyce, Lynn Hunt, and Margaret Jacob. 1994. Telling the Truth About
History. New York: Norton.
Barton, Allen H. 2001. “Paul Lazarsfeld as Institutional Inventor.” Interna-
tional Journal of Public Opinion Research 13: 245–269.
Berinsky, Adam J. 2006. “American Public Opinion in the 1930s and 1940s:
The Analysis of Quota-Controlled Sample Survey Data.” Public Opinion
Quarterly 70: 499–529.
Bryson, Maurice C. 1976. “The Literary Digest Poll: The Making of a Statis-
tical Myth.” The American Statistician 30: 184–185.
Burton, Antoinette, ed. 2005. Archive Stories: Facts, Fictions, and the Writing
of History. Durham, NC: Duke University Press.
Cahalan, Don. 1989. “Comment: The Digest Poll Rides Again!” Public
Opinion Quarterly 53: 129–133.
Converse, Jean M. 1987. Survey Research in the United States: Roots and
Emergence 1890–1960. Berkeley: University of California Press.
Eisinger, Robert. 2003. The Evolution of Presidential Polling. New York:
Foucault, Michel. 1982. The Archeology of Knowledge. New York: Vintage.
Fried, Amy. 1997. Muffled Echoes: Oliver North and the Politics of Public
Opinion. New York: Columbia University Press.
———. 2006. “The Forgotten Lindsay Rogers and the Development of Amer-
ican Political Science.” American Political Science Review 100 (4): 555–
561.
———. 2012. Pathways to Polling: Crisis, Cooperation and the Making of
Public Opinion Professions. New York: Routledge.
Fried, Amy, and Douglas B. Harris. 2010. “Governing with the Polls.” The
Historian 72(2): 321–353.
Gallup, George. 1939. Public Opinion in a Democracy. Princeton, NJ:

Princeton University Press.
Gallup, George, and Saul Forbes Rae. 1940. The Pulse of Democracy. New
York: Simon and Schuster.
Ginsberg, Benjamin. 1986. The Captive Public. New York: Basic Books.
Glaser, Barney G., and Strauss, Anselm L. 1967. The Discovery of Grounded
Theory: Strategies for Qualitative Research. Chicago: Aldine Publishing
Company.
Harris, Douglas B. 1998. “The Rise of the Public Speakership.” Political
Science Quarterly 113: 193–212.
———. 2005. “House Majority Party Leaders’ Uses of Public Opinion Infor-
mation.” Congress & the Presidency 32(2): 133–155.
Heith, Diane. 2004. Polling to Govern. Palo Alto, CA: Stanford University
Press.
Herbst, Susan. 1993. Numbered Voices: How Opinion Polling Has Shaped
American Politics. Chicago: University of Chicago Press.
Igo, Sarah. 2007. The Averaged American: Surveys, Citizens, and the Making
of a Mass Public. Cambridge, MA: Harvard University Press.
Jacobs, Lawrence R., and Robert Y. Shapiro. 1994. “Issues, Candidate Image,
and Priming: The Use of Private Polls in Kennedy’s 1960 Presidential
Campaign.” American Political Science Review 88: 527–540.
———. 1995. “The Rise of Presidential Polling: The Nixon White House in
Historical Perspective.” Public Opinion Quarterly 59: 163–195.
———. 2000. Politicians Don’t Pander. Chicago: University of Chicago
Press.
Karabell, Zachary. 2000. The Last Campaign: How Harry Truman Won the
1948 Election. New York: Knopf.
Korzi, Michael J. 2000. “Lapsed Memory? The Roots of American Public
Opinion Research.” Polity 33: 49–75.
Lazarsfeld, Paul F. 1982. “An Episode in the History of Social Research: A
Memoir.” In The Varied Sociology of Paul F. Lazarsfeld, edited by Patricia
L. Kendall, 11–73. New York: Columbia University Press.
Lee, Taeku. 2002. Mobilizing Public Opinion: Black Insurgency and Racial
Attitudes in the Civil Rights Era. Chicago: University of Chicago Press.
Lieberman, Robert C. 2002. “Ideas, Institutions, and Political Order:

Explaining Political Order.” American Political Science Review 96: 697–
712.
Likert, Rensis. 1948. “Opinion Studies and Government Policy.” Proceedings
of the American Philosophical Society 92: 341–350.
Moore, D. W. 1992. The Superpollsters: How They Measure and Manipulate
Public Opinion in America. New York: Four Walls Eight Windows.
Mosteller, Frederick, Herbert Hyman, Philip McCarthy, Eli Marks, and David
Truman. 1949. The Pre-Election Polls of 1948. New York: Social Science
Research Council.
Orren, Karen, and Stephen Skowronek. 2004. The Search for American Polit-
ical Development. Cambridge: Cambridge University Press.
Pierson, Paul. 2000a. “Increasing Returns, Path Dependence, and the Study
of Politics.” American Political Science Review 94: 251–267.
———. 2000b. “The Limits of Design: Explaining Institutional Origins and
Change.” Governance 13: 475–499.
Robinson, Daniel. 1999. The Measure of Democracy: Polling, Market
Research and Public Life, 1939–1945. Toronto: University of Toronto
Press.
Rogers, Lindsay. 1949. The Pollsters: Public Opinion, Politics and Democ-
ratic Leadership. New York: A. A. Knopf.
Rottinghaus, Brandon. 2006. “‘Dear Mr. President’: The Institutionalization
and Politicization of Public Opinion Mail in the White House.” Political
Science Quarterly 121(3): 451–476.
Rubin, Ernest. 1956. “Questions and Answers—Defective Samples.” The
American Statistician 10: 15–16.
Skott, Hans E. 1943. “Attitude Research in the Department of Agriculture.”
Public Opinion Quarterly 7: 280–292.
Skowronek, Stephen. 1982. Building a New American State. Cambridge:
Somit, Albert, and Joseph Tanenhaus. 1967. The Development of Political
Science: From Burgess to Behavioralism. Boston: Allyn and Bacon.
Squire, Peveril. 1988. “Why the 1936 Literary Digest Poll Failed.” Public
Opinion Quarterly 52: 125–133.
Stanton, Frank. 1991. The Reminiscences of Frank Stanton. Oral History

Research Office, Columbia University.
Thelen, David. 1996. Becoming Citizens in the Age of Television: How Ameri-
cans Challenged the Media and Seized Political Initiative During the Iran-
contra Debate. Chicago: University of Chicago Press.
Chapter 9
Digging�Through�Trash
Finding�What�You�Need�in�Municipal�Archives
Kathleen Sullivan and

Patricia Strach
As American cities grew in the second half of the nineteenth century,

so did their problems: fires consumed crowded housing, disease spread
through dirty water, and trash filled the streets. The banks of the Missis-
sippi River in New Orleans, for example, were “made a common recep-
tacle of, and reeking with garbage and filth of all kinds” (Sanitary
Commission of New Orleans 1854, 390), leading some observers to
proclaim it “the filthiest hole in the land” (Humphreys 1999, 54). The old
practices that worked for rural communities (burying garbage in the yard
or feeding it to pigs) no longer worked in America’s burgeoning munici-
palities. Though it was altogether clear that somebody had to do something
to make cities habitable, it was not at all certain who (government, private
actors, civic entities) or how (by what authority and with what resources).
Nineteenth-century municipalities afford scholars the opportunity to

see how fundamental problems—clean air, water, streets, and homes—
that must be addressed become government problems; how governments
muster the authority to take care of them; how governments negotiate the
resources they will need to do so; and how they build capacity in adminis-
trative structures. In describing the processes by which municipal govern-
ments across the country take on new problems, we have found that they
do so in different ways. But, if the problems are largely similar (dirty
water, air, and streets), then variation in the solutions raises the causal
question, Why?
The answers, it turns out, cannot be gleaned from quantitative data, offi-
cial ordinances, or secondary sources alone. Instead, they require trips to
municipal archives around the country to recreate the politics of the time,
the decisions of key leaders, the availability of resources, and the paths
and options that were not chosen. We illustrate the questions researchers
can ask, the answers they can construct, and the broader political general-
izations that can be made through our own research in nineteenth-century
garbage collection. In this chapter, we show that municipal archives offer
an important lens to study questions of politics and governance.
Even a seemingly straightforward task such as garbage collection

requires an administrative authority and capacity that many cities were
still struggling to develop in the nineteenth century. Picking up garbage
entailed finding and employing carts, horses, drivers, collectors, and
receptacles. Households and businesses needed to learn what they had to
do and then comply by putting the garbage out at the appointed time in
appropriate receptacles for efficient collection. Once it was collected, the
garbage needed to go somewhere where it could be processed—dumped
in a designated lot, incinerated, reduced, shipped out to sea or a river, fed
to swine—without producing a new health hazard for the citizenry.
Providing for garbage pickup and disposal was not sufficient; a city also
needed to ensure local compliance and enforcement. In New Orleans, cart
drivers were using kitchen garbage and dead animals to fill the streets,
inviting hogs to tear the streets apart to recover them (“Streets Filled with
Waste” 1878). In Pittsburgh, even though the garbage ordinance had been
in operation for decades, officials found that “suitable garbage cans with
Digging Through Trash 277
covers are frequently lacking in the houses.”1 Health officials in nearby

Allegheny noted that animal, vegetable, and fecal waste lay outside houses
“until the arrival of the cart or open wagon in which it is to be loaded and
taken over its route, giving off its vile smell until the place of dumping
is reached.”2 Despite measures to cover cans and carts, travelers to Pitts-
burgh took note that the stench from an uncovered garbage wagon was so
great that “it nauseated the passengers when the car was still half a block
away.”3 To solve public problems, cities need authority, capacity, and the
power of enforcement.
Although garbage is not a central object of study in political science,
it does address fundamental issues of governing. Like other seemingly
mundane local government concerns—water, sewers, streets, buildings,
and nuisances—garbage collection is essential to communities and repre-
sents the majority of what governments in America actually do. Further,
looking at local governments adds a comparative perspective speaking to
broader questions about how governments work, rather than just how the
American federal government works. We settled on municipal garbage
collection it its formative years, which began falteringly in the post–Civil
War period and reached a national trend in the 1890s and early twentieth
century.
A return to early garbage collection also provides a lens into current

political controversies. Because of the scope and immediacy of garbage
collection, cities relied on readily available resources, which included
private companies, and so it was not uncommon for garbage collection to
be conducted through a public-private partnership. Debates about public
and private governance today are often so charged because “contracting
out” is equated with trends toward privatization. By looking back through
history, we were able to extract our questions about the roles of public
and private actors from current hot-button political debates. From this
perspective, we were able to view the long-term practices of public and
private networks and to view them in context, to capture the social and
economic conditions as well as any political corruption of the day. Offi-
cials’ private papers, organizations’ meeting records, and other collections

offered insider accounts of the political opportunities and conflicts among
various interests. Such context allowed us to study political development
from a basic perspective—we were able to appreciate the garbage problem
as an imminent challenge and could trace the claims to political authority
that were made by those interests that initially promoted garbage collec-
tion. We were able to gauge the resources that were available to deci-
sion makers and the political and economic opportunities that presented
themselves to the problem solvers. Agency reports, newspaper coverage,
and registered complaints provide a perspective into how well the provi-
sions for garbage collection were enforced. Archival sources allowed us
to observe the generation of authority, capacity, and enforcement. The
various ways in which cities met these imperatives of governing allowed
us to conduct comparative studies of political development.
In sum, by studying nineteenth-century municipal garbage collection,

we were looking at a new problem for cities and a genuine question about
how to address it. Variation across cities told us that there was no one
“right” answer. Although quantitative data and secondary sources allowed
us to see which cities collected trash in which manner, they could not shed
light on how governments came to the answers they did or why they did
so. For these questions, we turned to the archives.
Essential�Considerations:�Local�Context
Urban growth in the nineteenth century offered the right time to study
and garbage the right issue, though the big question remained: What cities
would we study? There are many criteria that go into choosing historical
cases. First are questions of research design: Will one look at most similar
cases, illustrative cases, or outliers? Each of these different strategies may
point to a different locale. However, when doing archival research (espe-
cially in local archives) there are also practical considerations to take
into account. Is material available? Fires, floods, and purges have wiped
out records from certain cities at particular times. Trying to find govern-
ment records in the South, for example, is hampered by large gaps in
the Civil War period. Even if there are records, are they accessible? Just
because there are documents does not mean they are organized in any
way. Researchers may have a hard time wading through unopened boxes
of material. Further, records may be housed in places that are hard to get
to, have limited hours, or require a researcher to have special permission
(which he or she may not have).
In choosing our cases, we relied on aggregate data to help us find
comparable cities. During the Progressive Era, health professionals and
engineers turned to the task of cleaning up cities—advocating better
drainage, street construction, and sewers. These “Sanitarians” conducted
surveys of cities and kept detailed records. We found two sources to be
particularly helpful: George Waring’s (1887) two volumes in the Report
on the Social Statistics of Cities and Charles V. Chapin’s (1901) 970-page
book, Municipal Sanitation in the United States. The latter lists cities and
their trash collection practices. From the work of these researchers, we
were able to collect data about trash collection. Because we wanted to
know not just how cities collected trash but also why they were doing so in
different ways, we added historical data from the 1900 census of cities and
information about methods of disposal (Winslow and Hansen 1903). We
put all three types of data (trash collection, city information, and methods
of disposal) into a spreadsheet for further analysis.
These data showed that American municipalities were evenly split

in their trash collection practices at the time: one-third had collection
by municipal employees (“public” collection), one-third used private
contractors (“private” collection), and one-third had no municipal collec-
tion (“none”). There are a number of explanations that might account
for these differences: maybe the variation was due to city size, industrial
capacity, or proximity to waterways (and their ability to spread disease).
At the very least, we had variation, but it turned out we had a lot more.
These numbers masked important regional differences. Cities in the West
by and large had no municipal collection, which is not that surprising

given the region’s late development. (What “no collection” meant for
growing urban populations such as San Francisco’s, however, remained a
mystery to us.) Cities in the North, by contrast, most often relied on private
collection, whereas cities in the Midwest (sandwiched between these two
regions) struck a balance between private collection and no collection. But
we were most surprised by Southern cities, which overwhelmingly relied
on public collection.
Surely, political scientists know the South is exceptional; a look at any
journal will turn up regressions with a dummy variable for “South.” But
the direction of Southern exceptionalism in this case—where Southern
cities were doing more by government—went against most of what we
were led to expect. How could the economically depressed South provide
municipal services by a method that was premised on developed admin-
istrative structures while the industrialized and advanced North turned to
contractors? What role did strong political machines play in determining
collection methods?
Surely, someone must have an answer. We checked to see if other

scholars had mapped the territory. We looked through all of the rele-
vant, tangentially relevant, and potentially relevant secondary literature by
political scientists, historians, economists, sociologists, and public policy
scholars we could find. But there were no clear answers. We had found
a true puzzle, and we would have to conduct research into the particular
histories of cities to find the answer.
City selection would be important. We based our selection criteria

on what we had learned from the literature. Because city size mattered
for trash collection practices (small cities had the luxury of potentially
taking no action), we chose to examine large cities. Further, we wanted
comparable cities. We eliminated outliers such as New York and Chicago
—extreme population centers with unusual politics. We selected three
cities, one for each region: San Francisco (no collection, West), Pittsburgh
(private contractor, North), and New Orleans (public employee, South).
At this point, our readers may be thinking the same thing as many of our
colleagues—New Orleans is not a “typical” Southern city. Yes, we agree
wholeheartedly. But, we found that there was no typical Southern city
at this time. Baltimore was gigantic (with more than 500,000 people) by
1900 standards; General William Tecumseh Sherman had burned Atlanta
to the ground; Richmond was the seat of the Confederate government;
Washington, DC, was an outlier in how it collected trash and in that it
was the seat of the federal government; and Macon, Norfolk, Tampa, and
Dallas were too small to be comparable with our other cases. New Orleans
was the most comparable with Pittsburgh and San Francisco and as atyp-
ical as every other Southern city of the time.
By now, the reader is likely familiar with the challenges and rewards
of conducting archival research. We would like to suggest that both are
magnified when one is conducting research in municipal archives. Unlike
national offices—for which there are some standard rules of thumb—
municipalities are all over the map in (1) what they keep, (2) how they
keep it, (3) where they keep it, and (4) whether or not the material survived
political purges or natural disasters. At some archives, we were met with
attentive archivists, well-kept research space, and perfectly preserved
records. At others, staff did not think their collections could help us and
did not have dedicated space for us to work in, and we literally dusted off
records that had not been opened in a century. Both kinds of collections
had materials that were essential to our project. Our advice is to be neither
encouraged nor discouraged by the outward appearance of collections; the
substance is what matters.
When working with local collections, we cannot emphasize enough

the importance of local context. Municipal archives are often (though
certainly not always) in municipal buildings. Researchers ought to famil-
iarize themselves with local holidays and other events that may interfere
with scheduled hours. We learned this lesson the hard way. In one city,
we had carefully planned our schedule and carved out specific days for
specific archives. We looked online for hours and holidays. However, we
had not anticipated that a collection in a private building would close early
for a donor’s reception (to which we were not invited!). In New Orleans,
we planned a trip for the week before Mardi Gras, or so we thought. We
had not expected that Mardi Gras would actually begin early and public
buildings would close a day and half in advance of what we anticipated.
All of this is to say that websites can be helpful, but they do not spell out
what may be common knowledge to people who are from the area, and it
is worth calling to be sure.
Local context matters in another way: preparing and interpreting the
archival documents. Before going to any municipality, it is worth familiar-
izing oneself with the history and key players of the area. Certainly, histo-
rians steep themselves in single case studies. For political scientists, the
goal is not to tell the history of any particular location but to employ cases
to learn something about politics and governing more broadly. We relied
on secondary literature of cities to sort out the different political regimes,
the major political figures, and pertinent third parties such as civic groups
and business leaders, as well as general historical episodes such as disease
outbreaks and economic depressions and panics. We created detailed time
lines and a list of key players—a “who’s who” of local city politics, busi-
ness, and civic society. We made these time lines from those secondary
histories, original documents we could locate online (we found Google
Books to be a good resource for these during our time period), online
newspapers in the 19th Century U.S. Newspaper collection, and periodi-
cals that were available on Google Books. Our time lines for each city ran
roughly twenty-five to forty pages, and we estimate we invested ten hours
preparing for every one hour we spent in the archives.
Locating�the�Collections
With our cities selected and background research underway, we needed
to locate the libraries and archives that held municipal records. We used
a variety of approaches to familiarize ourselves with each city’s holdings

and to identify the collections.
The municipal government website is a good place to start. Although
a typical city website is geared toward current constituency concerns, it
should direct the reader to the city’s archives. The reader might be sent to
a phone number for the archivist, or the link might contain a searchable
catalog or a finding aid listing the archive’s holdings. Municipal archives
can be expected to contain minutes of the city council, volumes of munic-
ipal ordinances, annual reports of executive agencies, ledgers of agen-
cies, petitions, licenses, correspondence, and other related records. It all
depends on what records that city has decided or been able to retain. These
records, then, show both the nitty-gritty politics of cities as they develop
over time as well as the concrete actions that municipal agencies under-
took to meet the needs of their changing populations.
Municipal archives are housed in various locations. Some cities may

hold the archives in their city hall or in a separate municipal building.
The archives may have been transferred to the main branch of the public
library, as they were in San Francisco and New Orleans. Or, they may have
been transferred to a private library, as in the case of Pittsburgh, where the
city records are housed in the Senator John Heinz History Center. The city
website should direct the reader to the proper repository.
The city is not the only repository for public records. In cases in which
city records have been published, as with annual reports or municipal ordi-
nances, municipal records might be found at university libraries, histor-
ical societies, or state libraries. The university or state library website
might offer a guide to local resources. A “local government” section might
provide links to the local libraries and historical societies with holdings on
local government. Understanding municipal government, like other levels
of government, often entails looking outside of official government hold-
ings. Given our topic, our primary concern was with actual city records,
but we also found that the University of Pittsburgh archives in Pittsburgh
and the New Orleans collection at the Williams Research Center in New
Orleans provided additional municipal materials as well as collections

containing the records of relevant third-party actors, such as Department
of Health reports and records of civic associations.
Once we had a list of libraries and other institutions, we did a catalog

search of each of them. We cast a wide net, searching for terms such as
garbage, department of public works as well as other related agencies and
local civic associations, and other terms that might be garbage-related. Our
game plan was to access the records of public agencies, civic organiza-
tions, and third-party reports. Our catalog searches produced records of the
Civic Club of Allegheny County, records of an engineering company, and
the papers of a political boss at the University of Pittsburgh. The Senator
John Heinz History Center is a privately owned center with a floor devoted
to a library and archives. It contains city records, including annual reports
of city agencies. In New Orleans, the city records are available at the main
public library. We supplemented its holdings with a visit to the Historic
New Orleans Collection at the Williams Research Center, which holds
private papers of a civic leader and the papers of a local sanitary organi-
zation.
Because we were working together long-distance, we started a Google

Doc for each city we planned to visit. We listed each library, posting its
address, hours, and policies. We then copied and pasted each of the items
or collections we wanted to view. By having all of this information in one
document, we were able to schedule our trips by prioritizing the items we
wanted to view and the libraries we wanted to visit and by gauging how
much time we would need to spend in each library.
It does not hurt to double-check that list against a search in Google

Books, ArchivesUSA, or WorldCat. If items are available digitally, then it
would be wasteful of precious time during the visit to read it in an archive.
That said, digital resources should be used carefully. Handwritten docu-
ments are best read in the original. Even typeset documents might be only
partially available on Google Books, in which case it is still worth it to
visit the original document. In our experience, Google Books offers some
municipal materials online, but it is seldom comprehensive in presenting

all of the annual reports. Thus, online resources have served as a supple-
ment to research rather than a replacement for archival visits. Online
resources have been useful for offering a preliminary view of a city’s docu-
ments, showing us how they are organized, what is available, and what
we might expect when we get there.
Secondary literature on the history of our respective cities proved

crucial here. Different cities housed their garbage collection in different
agencies, so secondary literature helped us to identify the classifications
and terms that were used in the particular cities we visited. Even then, it
took some sorting out to locate the proper department. Current garbage
collection in Pittsburgh is organized in the Department of Public Works,
but we did not find garbage mentioned in that department in the later nine-
teenth century because collection was initially administered through the
Department of Public Safety. Because of such head-scratching moments,
it is better to cast a wide net in the initial search for holdings at the respec-
tive libraries. When we hit a snag and thought that there was no record
of garbage in Pittsburgh, we returned to our Google Doc for other terms,
then consulted the Department of Public Safety records and found plenty
of municipal activity on garbage.
Gaining�Access
Overall, we found it was not hard to gain access to municipal archives.
In part, this was because we were not looking at anyone’s personal
papers. Unlike the papers of national elites—such as members of Congress
or political party leaders—municipal papers do not offer the chance to
embarrass a particular person or leak national secrets. We never found
records redacted. In fact, we found just the opposite: local government
papers showing graft and corruption!
Though federal offices, such as Congress, have standard record collec-

tion regulations and practices, we have found that cities do not have stan-
dard record management.4 Instead, how documents are collected, which

documents are kept, and where they are located varies not only across
cities but also over time. If a municipality lacked formal regulations for
depositing public records, then the researcher is left with the vagaries of
local practices.
Special�Challenges
Municipal records from the late nineteenth and early twentieth centuries
are infrequently accessed in municipal archives. Genealogical researchers
are more likely to access records that can be searched by name, and only
a few historians are looking at agency records. For these reasons, even
the librarians may not be familiar with the contents of the holdings. One
librarian told us that we would not find anything in the volumes of one
of the agencies. We said we would look anyway, and we found a wealth
of data on garbage. The lesson is that it does not hurt to look; it is better
to pursue a dead end than to miss out on a source of information. The
challenges in studying underutilized sources also present opportunities.
One may find oneself asking for help in cutting the pages of a book that
has never been opened or being the first patron to open a set of correspon-
dence.
Deciphering handwriting and abbreviations presents another challenge,

one that we encountered each time we visited a new city or each time a
city underwent a regime change and adopted a new style of note taking
and recording. For example, we spent days reading through payrolls for
the New Orleans Department of Public Works (see figure 6) before we
were struck by the odd pattern of names. Some had first names or initials
preceding their last names, but others had simply “do” or hatch marks.
Going back and looking through the lists again, we discovered that the
“do” or ditto marks that filled the columns were under the title “Mrs.” We
were looking at a list of names of women who were receiving payment
from the city, which tipped us off to some significant gendered policies in
New Orleans and a separate paper.5
Figure 6. Payroll, Department of Public Works, 1883.
Source. City Archives, New Orleans Public Library.

The ledgers and correspondence that we found in the archives also

provided some of the answers to our questions about how an econom-
ically depressed South could provide municipal services. Letters from
the commissioner of public works about “intoxication, negligence, and
disobedience” among the laborers on the books illustrated the operations
of public employment.6 Though the initial statistics we found on Southern
public employees puzzled us, archival sources indicated that public collec-
tion was no indication of adequate or well-organized collection.
Researchers who do archival research in a single city may be able to
devote a significant amount of time to familiarizing themselves with city
history. Were that the case for us, we would have planned long visits and
taken our time with the documents. But, because we needed to collect
particular material from a number of cities at the same time as we had
obligations back home (such as teaching), we were faced with short
research visits in each city. We had to balance the need to visit multiple
cities with the imperative of doing particular research in each city. Other
researchers may likewise have to make similar trade-offs. We decided to
allot less than a week for each research trip, spending about four business
days in each city. We needed to move fast. Digital camera technology is
good enough that we could bring our personal point-and-shoot cameras
and snap photos of documents. This was especially helpful with annual
reports and ledgers.7
The ease of digital photography allows researchers to move through

documents quickly. But one should not be too hasty. Take the time to get
to know the documents, to understand how they are organized, to grasp
the logic of the record keeper. By familiarizing oneself with the document,
one can know what to expect, what is missing, and what else should be
explored. For example, the way things look matters. In New Orleans, a
single administrator in the first decade of the twentieth century tried to
professionalize his office. We did not comprehend the politics right away,
but we took note and filled in details later with secondary literature. In
that case, a change in record keeping alerted us to larger political changes.
Interpreting�the�Archival�Record
Municipal archives offer the chance to see inside hundreds of local
governments, but this comes with the obligation to do one’s homework to
understand the context of local government. For political scientists, there
is a delicate balancing act between spending scarce time in the archives
collecting data and interpreting that data. We have no rule of thumb
about the “right” balance between the two. Instead, we offer the following
advice. First, spend a few moments with the documents, reflecting on
what they are, what they contain, and why they have been collected and
preserved. This exercise offers some insight into what one is collecting
and why. For example, after poring through the archives finding references
to trash frustratingly scattered throughout various municipal volumes, we
suddenly found them collected and reported specifically. It would have
been all too easy to let out a sigh of relief and say “thank goodness!”
However, we found it more helpful (even if it was more work) to think
about what had changed that caused administrators to start collecting these
data, reporting these data, and saving them. Second, our time lines and
“who’s who” lists were essential to understanding the written documents
while we were in the archives. In fact, we continued to add to them as
we worked (we recommend bringing both paper and electronic copies;
some archives permit laptops but not paper, others allow paper but not
electronic equipment). Third, though the archives may be open for eight
hours, archival research trips often involve fourteen- to sixteen-hour days.
After the archives closed, we would stop and eat and discuss what we
had found and puzzle over the new questions that the material raised and
the old questions that we sought to answer. Evenings are a good time to
continue to interpret the record. Taking time to review helps to outline the
best course of action for the next day. Because we were moving through
the documents so quickly during the day, we needed the evenings to orga-
nize our information, figure out what we had seen during the day, conduct
searches to answer questions that had arisen, and add any new searches
to our list for the next day.
We used the Internet and our ace research assistant to try to find answers
(through secondary sources, online municipal records, newspapers, and
magazines, or even what additional archives we might wish to search
next). We also “processed” the hundreds of documents we collected during
the day. After the archives closed, we took our digital images, uploaded
them to our computer, and used a program (Picasa is our current favorite)
to label them (year, administrative agency, source, regarding) and mark
our favorites so they are easily searchable and rearranged. Because there
were two of us working on this project, we liked storing them online so we
both could have immediate access to them. Programs (such as Picasa or
Photoshop) can also be used to clean documents up, enlarge difficult-to-
read words, and brighten up dark copies. We looked over them to see what
we had collected, how it could answer our questions, and identify what
we might have neglected. Did we accidentally skip a year or a depart-
ment? Did we find something useful in later volumes that we needed to
go back and collect for earlier years? Were our photographs legible? Of
course, with so many pages, we could not thoroughly read them all; this
was more of an overview for us to circumvent the avoidable situation of
getting home only to find we did not actually get what we needed.
Roads�Not�Taken
Municipal archives have a wealth of information, not just about trash,
but about how governments operate. Looking at ordinances, city records,
political memos, and personal papers allows researchers to reconstruct
how governments have developed over time. One road we have not (yet)
taken is to ask why the federal government did not take up garbage collec-
tion. After all, a projection of political development is for programs to start
at the state and local levels and then get taken up by the federal govern-
ment. Garbage collection has remained local, yet the federal government
had a brief flirtation with garbage collection during World War I, when
it experimented with feeding garbage to hogs. The federal government’s
study of garbage offers a chance to examine why it decides to take up
some local projects but not others.8 Regardless of one’s interests, the infor-
mation in these archives could launch countless dissertations, books, and
research projects. To give a sense of the range of possible questions and
the ways to go about answering them, we highlight two recent pieces by
scholars using municipal archives.
Municipal archives can be used to explore what may seem like histor-
ical anomalies to offer broader lessons about politics. Jessica Wang
(2012) did just this when she detailed an interesting empirical puzzle.
In the mid-nineteenth century, the American Society for the Prevention
of Cruelty to Animals (ASPCA) was charged with enforcing New York
City’s animal control ordinances. Why was a nongovernmental organi-
zation acting as executive enforcement? Wang documented the power
wielded by the ASPCA and its ongoing confrontations with the New York
City Department of Health. Sources include Department/Board of Health
annual reports, ASPCA publications, and city newspapers. By tracking
the success of animal control efforts and the struggle for power between
a city agency and a voluntary association, Wang drew larger lessons on
public-private relations in governing.
Looking at cities, however, also allows a comparative perspective

across different American governments. One of the great benefits of
subnational research (and municipal research in particular) is the large
number of cases from which scholars can draw. Jessica Trounstine exem-
plified this approach in Political Monopolies in American Cities (2008).
She showed that political monopolies, whether they are machines or
reformers, bias the political system and reduce competition. Working on
behalf of a much narrower constituency, both machines and reformers
thwart representative government (Trounstine 2008). To make such a
claim, Trounstine amassed both aggregate data on cities and their elected
officials and case studies on nine specific localities and performed an
in-depth analysis of two (See also: Trounstine 2006, 2011). The mixed-
methods approach here relied on digging out local voting records.
Wang and Trounstine asked different questions, relied on different data,

and used different methods. But, by employing local records, they got to
questions at the heart of governing. Wang’s work starts with a very specific
historical event and draws lessons out from it, whereas Trounstine’s work
looks comparatively across many cities and drills down with individual
case studies. Though we will be the first to admit that national records
in federal agencies are often easier to access, better organized, and more
familiar, taking the time to investigate municipal questions by going to
archives offers distinct advantages.
First, municipal archives allow scholars to explore questions that are
largely ignored at the federal level. In our case, we were interested in
questions about the public and private resources governments used to
solve problems. There are only a limited number of policy areas at the
federal level that we could look at to answer these questions, and even here
they are not always comparable. But researching cities—which are faced
with similar problems at similar points in time—allowed us to leverage
both aggregate quantitative data (to provide a bird’s-eye view) as well as
specific comparative historical case studies.
Second, though scholars focus on the federal level, for most Americans
local governments are “close to home.” They deal with the immediate
concerns that many individuals face (from potholes to trash collection to
education). Throughout history, most Americans have been exposed to
local concerns first and foremost. Archives are filled with materials about
how governments have addressed these local concerns and how citizens
have reacted to both the concerns and governments’ attempts to address
them.
Third, using archives to study municipal governments, from a purely

descriptive standpoint, tells stories that are unfamiliar to most polit-
ical scientists. The basic facts of what governments did and how they
did it are not only fascinating, but they are unavailable through other
sources (secondary histories, quantitative data). Insights gleaned from
these studies can change the way political scientists think about how
governments have developed, why they have developed this way, and
what this reveals about government authority and capacity more generally.
Endnotes
1. Bureau of Municipal Research, The City of Pittsburgh, Pennsylvania,

Report on a Survey of the Department of Public Health, Department of
City Controller, City Council, Department of Public Works et al. (Pitts-
burgh: Pittsburgh Printing Co., 1913), 24. Historic Pittsburgh General
Text Collection http://digital.library.pitt.edu/cgi-bin/t/text/text-idx?idno
= 00avd2227m;view = toc;c = pitttext. Accessed July 27, 2011.
2. Pittsburgh annexed Allegheny in 1907. Health Officer’s Report, Annual
Report, City of Allegheny, 1880, 339. Library and Archives Division,
Historical Society of Western Pennsylvania, Pittsburgh, PA.
3. “Campaign for Cleaner Streets May Be Started by Civic Club,” East
Liberty Tribune, August 10, 1917, Civic Club of Allegheny County
Records, Archives Service Center, University of Pittsburgh.
4. For example, representatives, senators, and their staffs receive a packet
of materials concerning office management, record keeping, and best
archival practices during their orientations to Capitol Hill. Adhering to
these standards is purely voluntary, and there are still idiosyncrasies in
record keeping that are specific to each office. See Records Management
Manual for Members, available at http://www.archivists.org/saagroups/
cpr/publications/Manual%20for%20Members.pdf. Members also receive
free storage of their papers for up to one year upon leaving office and
free shipping of their papers to a domestic location of their choice at the
conclusion of that year.
5. Kathleen Sullivan and Patricia Strach, “Widows’ Carts: Race, Gender,
and Governing in Nineteenth Century New Orleans” (unpublished manu-
script).
6. Letter from George Smith to Ed Egan, dated January 7, 1909, Letters
and Orders of the General Superintendent of Public Works, New Orleans
Public Library.
7. Ask the archivist for policies on photographing. They will likely require
that flashes be turned off. Libraries differ on whether they allow tripods.
There are desktop tripods which are helpful both for steady shots and
for allowing for hands-free photographing. We did not use one, and most
of our shots came out fine. Be sure to keep meticulous records when
photographing. It is useful to snap a photo of the cover page or write up
a note with bibliographic information and take a shot of that at the start
of each new document.
8. See “Army Leads in Waste Prevention and Utilization,” July 4, 1918,
Engineering News-Record 81 (1): 25–26.
Bibliography
Chapin, Charles V. 1901. Municipal Sanitation in the United States. Provi-

dence, RI: Snow & Farnham.
Humphreys, Margaret. 1999. Yellow Fever and the South. Baltimore: Johns
Hopkins University Press.
New Orleans City Archives. Louisiana Division: City Archives and Special
Collections. New Orleans Public Library, New Orleans, Louisiana.
Sanitary Commission of New Orleans. 1854. Report of the Sanitary Commis-
sion of New Orleans on the Epidemic Yellow Fever, 390. New Orleans:
Picayune Office. Retrieved from Google Books (http://books.google.com)
on June 21, 2011.
“Streets Filled with Waste.” 1878. Vermont Watchman and State Journal,
November 13, supplement, 1.
Trounstine, Jessica. 2006. “Dominant Regimes and the Demise of Urban
Democracy.” Journal of Politics 68: 879–893.
———. 2008. Political Monopolies in American Cities: The Rise and Fall of
Bosses and Reformers. Chicago: Chicago University Press.
———. 2011. “Evidence of a Local Incumbency Advantage.” Legislative
Wang, Jessica. 2012. “Dogs and the Making of the American State: Voluntary
Association, State Power, and the Politics of Animal Control in New York
City, 1850–1920.” Journal of American History History 98: 998–1024.
Waring, George E. 1887. Report on the Social Statistics of Cities. Washington,
DC: Government Printing Office.
Winslow, C. E. A., and P. Hansen. 1903. “Some Statistics of Garbage Disposal
for the Larger American Cities in 1902.” Public Health Papers and
Reports 29: 141–165.
Chapter 10
In�Search�of�Influence
Archival�Research�and
the�Study�of�Interest�Groups
Tracy Roof
As I began my first semester of graduate school in Baltimore in the fall

of 1994, President Bill Clinton’s far-reaching proposal to guarantee every
American health insurance was finally pronounced dead. A couple of
months later, the Republicans won control of both the House and Senate
for the first time in forty years, pledging to roll back America’s limited
welfare state. Witnessing these events unfold, I became preoccupied with
the idea of American exceptionalism. Why had the United States proven
so impervious to the social democratic programs, such as universal health
care, that most other advanced, industrialized nations had adopted? As I
began to explore the literature addressing this issue, the role of organized
labor figured prominently in most explanations. The United States has a
more limited welfare state in part because the American labor movement
never created a viable labor party, never attained the level of political influ-
ence that was shared by labor movements in other Western countries, and
focused on collective bargaining rather than politics to improve workers’
standard of living. Much of this work was historical in nature, but even
in explaining the contemporary failure of Clinton’s health care policy as
well as the Republican takeover of Congress, pundits repeatedly pointed
to the declining power of organized labor as a factor. My research came
to focus on why organized labor had apparently exerted so little influence
over public policy.
Unions represented more than a third of the workforce coming out

of World War II, but organized labor failed to translate this numerical
strength into comparable political success. Only a few historians have
examined labor’s role in the postwar period, and even fewer have looked
at labor’s efforts to influence public policy. Political scientists such as J.
David Greenstone (1977) had largely found labor to be an influential actor
in Democratic politics, but they had not focused on why labor accom-
plished so few of its policy goals. Had labor pursued misguided polit-
ical strategies? Had it failed to mobilize its members? Was the movement
too internally divided among competing goals? Had it found powerful
politicians unresponsive to its demands? Although I conducted an initial
round of exploratory interviews with current labor activists, I quickly real-
ized that I needed to understand how labor’s contemporary struggles over
public policy—such as the battle for health care in the Clinton adminis-
tration—fit into the history of labor’s efforts to expand the welfare state.
Had labor ever lived up to its potential in pushing for the expansion of the
welfare state, and had labor’s power over public policy in fact declined
from some heyday in the past? I turned to archival materials to find these
answers.
Given that the AFL-CIO’s national archives were only forty-five

minutes away in Silver Spring, Maryland, I had the luxury of casually
perusing documents on the AFL-CIO’s political activities—particularly
its lobbying—without an overarching thesis in mind. Every day I would
return home with a folder full of photocopies on the blue paper the archives
use to distinguish copies from the original documents. A now-tattered
photocopy of a seven-page memo from 1966 written by Andrew Biemiller,
In Search of Influence 301
the AFL-CIO’s chief lobbyist and a former congressman from Wisconsin,

stood out and came to shape my argument, my theoretical approach,
and the rest of the research project. It was a detailed description of the
labor movement’s strategy and a chronology of the day-to-day events in
a fifteen-month battle over the repeal of a provision of the Taft-Hartley
Act known as 14(b). Taft-Hartley had been passed over President Harry
S. Truman’s veto by the Republican-controlled Congress in 1947 with the
goal of reining in the power organized labor had gained over the course
of the New Deal and World War II. The labor movement had tried unsuc-
cessfully to get the entire bill repealed when Democrats regained control
of Congress during the Truman administration. But by the 1960s, labor
had come to focus on repealing what was widely considered one of the
most odious provisions, section 14(b), which allowed states to pass “right-
to-work” laws. These laws prohibited the negotiation of union security
arrangements that made it easier for unions to solve collective action prob-
lems by requiring workers in unionized workplaces to join the union or
pay for their share of the costs of collective bargaining. Many Southern
and rural Western states had adopted such laws, which reinforced union
weakness in these states and contributed to a pattern of extreme regional
variation in union strength. Fights against proposed right-to-work laws
were also costly and diverted unions’ resources from national politics.
Thus, the repeal of 14(b) was one of labor’s top policy goals in the Lyndon
Johnson administration, and the failure of the proposal due to a Senate
filibuster was widely viewed as evidence of labor’s political weakness and
President Johnson’s lack of commitment to labor’s agenda, even at what
is often considered to be the height of labor’s political influence in the
Great Society years.
But Biemiller’s memo suggests a more complicated picture. As I later

noted in my book, the memo read like “a textbook case study of the
obstacle path that is the American legislative process” (Roof 2011, 101).
His chronology revealed a high level of cooperation among Democratic
congressional leaders, the Johnson administration, and labor lobbyists in
trying to get the bill past numerous potential institutional obstacles from
committee to floor consideration. At every stage of the process in the

House, the labor movement managed to do what was necessary to get the
bill passed, from getting it referred to the most favorable subcommittee,
to strong-arming a recalcitrant committee chairman to call up the bill,
to clearing its path to the floor by helping to pass a reinstatement of a
House procedure known as the “twenty-one-day rule,” which prevented
the conservative Rules Committee from bottling up legislation indefi-
nitely. The bill finally passed on the House floor in the summer of 1965
by a vote of 221 to 203, and President Johnson called AFL-CIO president
George Meany on the House floor to urge quick action in the Senate.
But the Senate posed greater challenges. The memo noted that in an
early August meeting between Biemiller, Meany, and Republican minority
leader Everett Dirksen, the senator from rural Illinois signaled his inten-
tion to propose an amendment to the 14(b) repeal, amending the Consti-
tution to invalidate a recent Supreme Court decision requiring the appor-
tionment of seats in both houses of bicameral state legislatures based on
population. Labor and most liberals were very supportive of the court’s
decision because the bias towards rural representation in many state legis-
latures had undermined the power of urban and typically more unionized
areas. Dirksen threatened to filibuster the 14(b) repeal unless labor agreed
to drop its opposition to the amendment. The memo noted that Dirksen
warned he would “use every weapon at his command in the fight ahead.”1
Conservative opponents stalled the bill in committee through extended
debate and amendments, but it was finally reported in September. Dirksen
threatened that if 14(b) was brought up, the Senate would be in session
until “the snow falls.”2 As I would discover in later research in the Lyndon
Johnson archives, Majority Leader Mike Mansfield was very skeptical
that cloture could be invoked, but he called up the bill anyway out of the
Democrats’ obligation to labor. After several failed cloture votes, Mans-
field recessed the Senate. But, under labor and administration pressure, the
bill was again taken up when the Senate came back into session in 1966.
In the final vote in February, fifty-one senators voted in favor of cloture,
with one pro-repeal senator out due to illness. Labor had majority support,
but nowhere near the sixty-seven-vote supermajority that was required to

invoke cloture at the time.
The memo drove home to me that the failure of 14(b) reflected more
about the high hurdles to policy change in America’s uniquely convoluted
legislative process than it did about the political weakness or failed strate-
gies of organized labor. The particulars of the legislative battle actually
suggested that labor was quite politically savvy and exerted a great deal
of influence in Congress and the Johnson administration. But this influ-
ence could not overcome minority obstruction. Defending the Johnson
administration and its commitment to labor in an oral history interview
some twenty years after the 14(b) fight, White House staffer Lawrence
O’Brien noted that passing the 14(b) repeal after Dirksen committed to
filibustering was simply “an impossible task.”3 This legislative episode,
as it was described in the Biemiller memo, suggested to me an entirely
different approach to judging the political influence of organized labor
that no scholar had clearly taken before. Similar to Amy Fried’s experi-
ence, described in chapter 8 in this volume, the archival research led me to
adopt a historical institutionalist approach in evaluating the role of orga-
nized labor in American public policy. The exceptional weakness of the
labor movement when compared with those of other industrialized coun-
tries was in part embedded in U.S. political institutions, particularly the
institutions of the legislative process which privilege minority defense of
the status quo, not just on labor law legislation but on a range of welfare
state policies.
With a guiding thesis in mind, I undertook additional research at the

archives of the UAW and CIO at Wayne State University and the presi-
dential archives of Harry Truman, Lyndon Johnson, and Jimmy Carter,
focusing on labor’s legislative efforts in three policy areas (labor law
reform, full employment, and health care) from the 1940s through the
1970s. I found similar patterns across the Democratic administrations and
documentary evidence that challenged contemporary accounts of labor’s
lack of political influence. For example, it was clear that the Carter admin-
istration actually worked quite hard to pass labor’s top legislative priority
of a package of labor law reforms and was confident until the final days
of the losing battle that a sixtieth vote could be found to invoke cloture
on a filibuster against the bill. But once my attention was trained on
the constraints imposed on labor’s political influence by legislative insti-
tutions such as the House Rules Committee and the Senate filibuster,
additional themes repeatedly appeared in the documents that led me to
expand the research project again. It became clear that labor leaders’
frustration with policy failures in the early postwar period led labor to
develop and pursue a fairly coherent political strategy over this thirty-year
period of obtaining what legislative gains it could while trying to realign
the Democratic party away from Southern conservatism towards urban
liberalism and to reform congressional rules and procedures to empower
the majority in the majority party in the legislative process. Documents
detailing labor’s lobbying of presidents, presidential candidates, members
of Congress, and Democratic Party officials suggested how highly civil
rights legislation was prioritized by national labor organizations from
the 1940s forward, with internal documents citing the enfranchisement
of black people as a major component in a strategy to undermine the
power of conservative Southern Democrats. Evidence of labor’s electoral
activities further showed a sustained effort to mobilize African American
voters, even in the South prior to the passage of major civil rights legis-
lation. Other documents showed how hard labor pressured Democrats,
including congressional leaders, administration officials, and Democratic
National Committee leaders, to take on various issues of congressional
reform, including reform of the Rules Committee, seniority provisions,
the cloture threshold, and the Democratic caucus. They also revealed how
proposals that labor helped to shape in the 1940s were eventually adopted.
These objectives took decades to achieve and had a number of unintended
consequences, but this was another arena of influence in which labor’s
political mobilization appeared far more consequential for the course of
American political development in the postwar period than most labor
scholars had acknowledged. These reforms reverberated in the political
system for decades, playing a central role in reshaping party alignments,

the legislative process, and contemporary U.S. politics, even as the union-
ized percentage of the workforce declined. Without the archival research,
I am not sure I would have picked up on the extent of labor’s role in these
areas.
The�Study�of�Organized�Interests
Much of the contemporary political science literature on interest groups
has focused on addressing three overarching concerns:
1. describing the contours of the interest group universe in terms of

the number and type of groups, as well as the tactics and strategies they
typically employ;
2. investigating the organizational development of groups, including

how they are formed, how they acquire and use resources, and how they
attract members and support; and
3. evaluating the influence of groups in electoral politics and the public
policy–making process.
Some studies, particularly those focused on the aggregate influence of

interest groups as a class of political actors, address only one of these ques-
tions, whereas others, particularly those that provide an in-depth analysis
of one organization or the range of groups that are active in one sector or
policy area, look at all three.
Political science is an eclectic discipline, and researchers have

employed a diverse range of methodological approaches in the study of
interest groups. Large-scale surveys have been used to assess the range
of groups that are active in politics, the origins of groups, and the strate-
gies groups use (Schlozman and Tierney 1986; Knoke 1990; Walker 1991;
Heinz et al. 1993; Gray and Lowery 1996). Much of the research on
groups’ involvement in elections has been based on quantitative analyses
of data that have become widely available in recent decades, such as

records of campaign contributions, public opinion polls, voter turnout,
and other electoral statistics. Other researchers have made creative use of
available data. For example, Jeffrey Berry (1999) used a series of content
analyses of references to groups in press reports, group appearances before
Congress, and categorizations of the issues addressed by legislation that
was the subject of congressional hearings to measure the changing influ-
ence of various categories of groups on the political agenda over a thirty-
year period. But the most common method of collecting information on
interest groups has been interviews with group representatives and partic-
ipants in the policy-making process. This technique has been used to
collect information on interest groups’ origins (both in studies of indi-
vidual groups, as in, for example, McFarland 1984, and in studies of
the interest group system, as in, for example, Walker 1991), their elec-
toral activities (see, for example, Francia 2006), their policy positions
(see, for example, Martin 2000), and their lobbying strategies (see, for
example, Kollman 1998; Goldstein 1999; and Nownes 2006). Interviews
with more than three hundred lobbyists and government officials on a
random sample of policy issues were also the basis of the path-breaking
study by Baumgartner and his colleagues of the impact of organized inter-
ests’ lobbying on policy debates and outcomes (Baumgartner et al. 2009).
Although knowledge of groups’ influence in the contemporary period has
expanded considerably in recent decades as a result of such research,
scholars’ reliance on these methods has necessarily limited the time hori-
zons of most interest group scholarship. Much of the political science liter-
ature, to the degree that it covers historical events, still relies on secondary
academic sources and contemporary press accounts. But this is changing.
A growing number of studies have utilized archival research, either

alone or in combination with methods such as interviews, to investigate
the operation and activities of interest groups in the past and changes
over time. As in my own research, most of these studies have been situ-
ated in the subfield of American political development. Although it is by
no means exhaustive, the following section will categorize some of the
most prominent strains in interest group research utilizing archives and

will provide an overview of the various types of questions and theoretical
issues archival research has been used to address.
Archives�and�Interest�Group�Research
In one of the few studies utilizing archival research that also attempts to
quantify the full range of groups active in Washington policy making prior
to World War II, Daniel Tichenor and Richard Harris called attention to the
need to “expand the time horizons” of interest group research beyond the
last half century to include the Progressive Era, when modern liberalism
began to “take shape” (2002, 588). Acknowledging that research into the
aggregate role of the universe of groups—as opposed to case studies of
particular groups—has been limited by the use of sources considered to
be reliable, such as directories of organizations active in Washington that
cover only a limited time period, Tichenor and Harris looked to new histor-
ical sources. Using records of congressional committee testimony by orga-
nizations from 1833 to 1917, they showed the steep growth in the number
and diversity of interest groups appearing before Congress during the
Progressive Era. Because of the limits of their data, scholars working in the
postwar period missed this growth and tended to assume that the transfor-
mation and diversification of the interest group universe that was observed
in the postwar period was unprecedented. Tichenor and Harris used their
findings to challenge the universality of theories developed in the postwar
period that the growth of government drives interest group formation
rather than the other way around, because these groups emerged before the
growth in government the groups sought. They supplemented this aggre-
gate information on the interest group system with a case study of immi-
gration policy which utilized archival research into the activities of pro-
and anti-immigration groups. The archival materials revealed the strate-
gies of these groups and the heavy emphasis they placed on providing
expertise to government commissions and congressional committees as
a means of exerting influence in the policy-making process. Although
much of the literature on interest groups had suggested that cozy iron
triangle relationships among clientele groups, congressional committees,
and executive agencies were dominant into the 1960s, Tichenor and Harris
found a diverse issue network operating in the highly visible policy area
of immigration policy before the turn of the twentieth century. Clearly,
the rise in public interest groups in the 1960s and 1970s was not a new
development in American politics but a return to earlier patterns from the
Progressive Era. Tichenor and Harris concluded that the roots of modern
interest group politics go deeper into history than most post–World War
II studies suggest.
Other studies also looking at the Progressive Era have used archival
research to explore the origins, maintenance, and strategies of particular
groups or categories of groups. These studies found archival evidence
suggesting that many actors viewed interest group mobilization as an
attractive alternative to working through political parties, consistent with
the Progressive Era reformers’ hostility to parties. Based on research in
various archives of labor, farm, and women’s organizations, Elisabeth
Clemens (1997) traced the emergence and organizational development
of citizens’ groups in these three sectors, concentrating on the states
of California, Washington, and Wisconsin. She found that these groups
created a “people’s lobby” that used tactics of popular mobilization in
order to circumvent the entrenched interests that were tied to the dominant
two-party system in shaping Progressive Era public policies. Focusing
on just one group, Cathie Jo Martin (2006) used National Association
of Manufacturers (NAM) convention proceedings and President William
McKinley’s papers to trace the early development of the NAM as an
effort by McKinley and his supporters to overcome the sectionalism of
the parties at the turn of the century and to mobilize industrialists in all
regions of the country, including the Democratic South, behind the Repub-
licans’ agenda. Her study also helps to explain a perplexing reversal in the
NAM’s policy positions. She found that after failing to achieve its legisla-
tive initiatives and losing membership, the organization shifted from a
vision of economic coordination to laissez-faire individualism in order to

attract more members.
Looking at the postwar period, a study by McGee Young (2010) also
deals with the organization and maintenance of groups but attempts to
develop more universal principles about the nature of interest group devel-
opment rather than the characteristics of the political system in a particular
era. Young used interviews and archival research to explore the origins
and organizational development of two pairs of groups: the National Small
Business Association (NSBA) and the National Federation of Independent
Business (NFIB), and the Sierra Club and National Resources Defense
Council (NRDC). He argued that aggregate studies that look at the influ-
ence of interest groups as a class of political actors obscure the impact
of particular groups and the importance of understanding which groups
have flourished in the system and why. He discovered that the organi-
zational entrepreneurs who found organizations shape the groups’ polit-
ical identities, their tactics, and their ties to other political actors. These
core features of the groups’ formative identities shape their prospects for
success and survival later on as organizations have to adapt to changing
political opportunities. This process has important consequences because
which groups thrive, such as the NFIB, and which groups decline, such
as the NSBA, in turn shape policy debates and outcomes as well as the
representativeness of the interest group system of societal interests. Young
emphasized that the outcome of the competition among groups is not fore-
ordained in a test of the survival of the fittest but is the result of a complex
path-dependent process of organizational adjustment to changing circum-
stances. The historical documents that are available from these groups
allowed him to piece together examples of this process.
Although all of these studies have linked the origin and maintenance
of groups to their policy stances, another strain of interest group scholar-
ship utilizing archival research has focused more explicitly on the evolu-
tion of various groups’ positions in a particular policy area over time.
Victoria Hattam (1993) used archives, platforms, proceedings, and publi-
cations from the Knights of Labor and the AFL to tie the rise of collective
bargaining and business unionism as the dominant organizational strategy
of the AFL in the late nineteenth century to its confrontations with the
courts in its efforts to obtain legislation protecting workers’ rights and to
changing conceptions of class from the Knights’ focus on producers to
the AFL’s focus on wage earners. Her study thus challenges conventional
notions that the AFL eschewed political action because of the conser-
vatism of its leaders or the lack of class consciousness among union
members. In another study of labor, Janice Fine and Tichenor (2009) used
interviews and research in the papers of labor, immigrant rights, and anti-
immigration groups to explain organized labor’s shifting policy positions
on immigration from the mid-1800s through the contemporary period.
Marie Gottschalk (2000) also combined research in labor and business
association archives with interviews with representatives of labor, busi-
ness, and public interest groups to trace the evolving positions of these
groups on health care reform in the post–World War II period. She used her
empirical study to argue that labor lost the opportunity to build an effec-
tive political coalition behind a national health care system by endorsing
reforms it thought business would accept. In each of these examples,
the historical documents from the interest groups’ archives allowed the
researchers to see how and why the groups’ public policy positions shifted
with far more detail than could have been reconstructed from contempo-
rary press accounts alone.
As is true of the larger field of interest group scholarship, most studies

utilizing archival research on interest group activities focus on the impact
of groups on policy development in a particular area. The empirical
evidence of group influence is often used to support larger arguments
regarding the trajectory of American political development. For example,
in her influential study of the mobilization of women’s and veterans’
groups in the late 1800s and early 1900s, Theda Skocpol (1992) used
archival research in the records of national-, state-, and local-level orga-
nizations to show how effective these groups were in mobilizing for
pensions and other protections for veterans and mothers. In the process,
these groups built a nascent welfare state that was quite different from
those focused on benefits for male wage earners, which developed in most
other Western countries in response to the pressure of organized labor.
This early welfare state had not received much scholarly attention prior
to Skocpol’s work because so few researchers had explored the goals and
activities of these local groups, which virtually required archival research.
In another study covering the development of immigration policy over
the entire history of the United States, Tichenor combined interviews with
extensive research into the archives of pro- and anti-immigrant groups
and activists, labor archives, presidential papers, congressional records,
and other government documents to look at how the “interactions between
political institutions, ideological traditions, and organized social inter-
ests” (2002, 5) encouraged the adoption of expansionist immigration poli-
cies at some points in American history and restrictive policies at others.
A range of studies on the historical development of public policies

also make a case for path dependency, providing evidence that interest
group strategies and policy decisions at one point in time shape the range
of policy options that are open to groups decades later. Several of these
studies look at groups’ role in the development of a mixed system of
public and private sector benefits for workers in the United States in
the post–World War II period. Jennifer Klein (2003) used research into
archives of various labor organizations, business organizations, insurance
companies, and federal agencies to trace the policy battles that encouraged
the growth of employer-provided health insurance and pension benefits
rather than expansive public programs. Jacob Hacker (2002) also used
archival research, primarily in the records of federal agencies, to explore
similar issues, including the evolution of the positions of business and
labor groups on health care and pension policy and the complementary
development of public and private sector programs. In another example,
Christy Ford Chapin (2010) used research in the Social Security Admin-
istration archives and the papers of leaders and activists in the American
Medical Association (AMA) and health insurance industry to illustrate
how the positions and activities of the AMA and the Health Insurance
Association of America (HIAA) shaped the American system of health

care financing. Two other examples are focused on civil rights policy. Paul
Frymer (2008) used the archives of the NAACP, various labor groups, and
federal agencies to examine the government’s efforts to integrate unions
and to describe how issues of class and race were fragmented into sepa-
rate policy spheres in policies regulating the workplace, in part because
of the strategies that were pursued by civil rights and labor organizations.
Anthony Chen (2009) also used research into the archives of members of
Congress and civil rights organizations to link the development of affir-
mative action to the outcome of earlier policy battles over fair employ-
ment practices legislation from the 1940s to the 1970s.
While also examining the impact of groups on public policy outcomes,

other studies have used archival evidence to establish patterns of influence
of interest groups from a given sector, such as agriculture and labor, in
various policy-making arenas over time. Hansen (1991) supplemented his
own interviews with group representatives and policy makers with oral
history interviews from various archives to tie the ebb and flow of agri-
cultural interest groups’ influence in Congress from 1919 to 1981 to the
groups’ ability to bolster members of Congress’s electoral prospects. Also
combining research at labor and presidential archives on labor’s activities
in the 1960s and 1970s with interviews of labor lobbyists who were active
in the 1980s and 1990s, Taylor Dark (1999) used evidence of labor groups’
influence in the executive branch, Congress, and Democratic Party struc-
tures to support his argument that labor’s political power did not decline
over this period but instead remained remarkably consistent.
Finally, archival research has also been used to explore the relationships
among various interest groups and organized constituencies. For example,
Alan Draper (1994) looked at the interaction of organized labor and the
civil rights movement in the South from 1954 to 1968 by utilizing research
in multiple labor archives and the papers of labor and civil rights leaders,
as well as interviews with labor activists, local- and state-level leaders,
and staff of labor organizations. His study makes it clear that the AFL-CIO
faced nearly insurmountable challenges in trying to maintain the support

of Southern union members while furthering its goal of progress on civil
rights. The relationship of labor to the civil rights movement in the South
had received very little media or scholarly attention, and it would have
been very difficult for Draper to have recreated this history without the
use of archival materials. In another example using both archival research
and interviews, Andrew Battista (2008) analyzed the relationship between
groups in the labor-liberal coalition from 1968 through the mid-2000s,
tracing the decline of the alliance and its revival, especially through the
activities of groups created by the more progressive wing of the labor
movement in the 1970s and 1980s, such as the Progressive Alliance and
the Economic Policy Institute. Very little scholarly attention had been
given to these groups, and the interviews and documentary materials were
necessary for Battista to find evidence for his study.
Advantages�and�Disadvantages�of�Archival�Research
The diverse range of studies employing archival research illustrates how
useful a tool it can be in studying interest groups. One of the key advan-
tages of archival research over other methodologies such as surveys, inter-
views, and quantitative analysis is that it allows the researcher to look at a
longer time horizon. Historical data may not be available, and participants
in past political events may simply no longer be around to interview or
survey. As a result, many of the studies reviewed earlier combine archival
research with other methodologies, particularly interviews, in order to
trace the evolution of group strategies and influence over time. More-
over, a researcher may not be able to secure interviews with some of the
most important players in political battles, such as the president. But the
researcher will have access to the president’s considerations through the
materials that are available in presidential archives. Archives also enable
researchers to study groups or issues that have not received much media
or scholarly attention and are therefore not covered in many secondary
sources, as in the works of Clemens, Martin, Young, Skocpol, Draper, and

Battista discussed earlier.
Archival research can also help the researcher get a more objective
perspective on a group’s political influence or internal operations. Politi-
cians and other political actors may have reasons to downplay the influ-
ence of unpopular interest groups or exaggerate their responsiveness to
important constituency groups. Group representatives may also be less
than forthcoming about internal divisions over policy positions, failures
to exert influence on policy makers’ decisions, or declining membership
or resources. Thus, public documents and contemporary press accounts
may be misleading, and group representatives or policy makers may have
reasons to present a subjective or skewed interpretation of events when
they are interviewed by a researcher. Archival research will often allow
a researcher to avoid these pitfalls because he or she has access to docu-
ments that were not intended for public consumption.
Archival materials also enable a researcher to gain a more comprehen-

sive understanding of a group’s policy positions by providing information
that may not be available anywhere else. A group’s archives (and often
information from other archives, such as those of members of Congress
or presidents) can provide information on the nuances of the group’s
changing positions over time and the strategies they used to accomplish
their policy goals. Archival materials can reveal internal deliberations and
disputes over these positions and strategies that were never made public.
Moreover, information may be uncovered on a group’s bottom line posi-
tion, as opposed to its ideal position that was typically presented in public.
A researcher will likely also get a better sense of how groups developed
and prioritized their agendas. Finally, archival materials may shed light on
groups’ efforts to build coalitions with other groups. As discussed earlier,
archival findings along these lines enriched the work of Martin, Young,
Fine and Tichenor, Gottschalk, Klein, Hacker, Chapin, and Battista.
Archival research affords the researcher insight into the myriad ways
that groups exert influence in the policy process. Political scientists often
judge a group as successful if the legislation it endorses becomes law.

But, if one simply looks at whether an organization has achieved its
overall policy goals or whether it has won or lost particular legisla-
tive battles, one may underestimate a group’s power. In contrast, if one
judges organizations’ influence by how responsive politicians are to its
requests, how much coordination there is between powerful politicians
and group leaders, and how much impact the groups have on particular
policy debates—regardless of whether the group is ultimately successful
in enacting its favored legislation—a much different picture emerges.
Archival materials that reveal the behind-the-scenes interactions between
policy makers and groups can shed light on these less-obvious forms of
influence. For example, in my own research, organized labor lost every
major labor law reform fight in Congress when the pro-labor position
was a change in the status quo since the passage of the National Labor
Relations Act in 1935, typically as the result of a Senate filibuster in
which labor had majority support. But labor exerted considerable influ-
ence in each of the battles. The studies by Tichenor, Hansen, and Dark that
were discussed earlier reflect similar insights, finding significant interest
group influence, even when groups did not prevail. By facilitating a more
comprehensive analysis than the measurement of win-loss ratios on legis-
lation, archival research can enable the researcher to more fully capture
the nature of interest group influence in Washington.
Archival research can also provide a much better window on groups’

roles in shaping politicians’ decisions in the broader context of the deci-
sion-making process. Groups’ demands are only one of many factors that
shape political outcomes, and archival materials can shed light on how
politicians weigh these multiple factors. In another example from my own
research, I found memos in the Truman archives documenting the volume
of mail urging Truman to veto the Taft-Hartley Act, much of it generated
by organized labor’s efforts. Other memos documented the overwhelm-
ingly critical tone of the newspaper editorial pages towards labor and
the negative tenor of public opinion. Letters from Southern members of
Congress warned against a veto, arguing it would destroy the Democratic
Party. However, highly confidential memos from labor relations special-

ists also revealed that the experts the administration consulted overwhelm-
ingly urged a veto of both bills out of fear their provisions would provoke
more industrial unrest. The documents made it clear that Truman was
aware he was potentially taking a costly stand with widespread conse-
quences. However, the political calculations on the veto were of most
interest, and they presaged Truman’s strategy going into the 1948 elec-
tions. As Clark Clifford noted in his famous memo on the factors shaping
the 1948 election, “As always, the South can be considered safely Demo-
cratic. And in formulating national policy, it can be safely ignored…[The
president] must, however, get along with the Westerners and with labor
if he is to be reelected.”4 In vetoing Taft-Hartley, Truman decided he had
more to gain than to lose from an alliance with labor, suggesting labor
had a great deal of influence within the administration and the national
Democratic Party. Many of the studies described earlier benefitted from
similar contributions from archival research.
Of course, there are also limitations and drawbacks in using archives

to study group influence. Many archives have embargoes on documents
of twenty to twenty-five years, and some individuals may contribute their
papers to an archive with a prohibition on their use while the contrib-
utor is still alive. Thus, a researcher may well be restricted to studying
events in the distant past. Although archival research is very useful for
tracing patterns of change and continuity and looking for universal prin-
ciples, if a researcher is studying a group’s influence over time, including
the contemporary period, he or she will be looking for patterns with very
different levels of information about events in the past and the present. It
may also be hard and time-consuming—or just plain impossible—to get
a complete and balanced view of a group’s influence. Ideally, one would
want to look at the archives of political allies as well as enemies and to
look at those of groups on all sides of an issue to mitigate the impact of the
subjectivity of the actors involved on one’s conclusions. But many groups
do not maintain archives. Moreover, archives can vary considerably in
their completeness, with many important records likely lost to time. As a
result, a researcher is not getting either a complete or a random sample

of information, which may be misleading. Another problem can be infor-
mation overload. I came back with thousands of pages of documents from
each of my presidential archives visits. It took months to review them all,
it was hard to catalog them, and many of the documents were ultimately
not very useful. It can take a lot of mining to find the occasional diamond.
Yet, when that happens, as with the memo I described at the opening of
this chapter, it can transform a research project.
Interest�Group�Archives�and�Advice�on�How�to�Use�Them
The use of archival research has likely been more common in the study of
the political activities of organized labor than in the study of other groups
and organized constituencies because the history of organized labor is
so well preserved. There are numerous collections of documents from
labor unions and labor organizations, including those of national-, state-,
and local-level organizations. The George Meany Memorial Archives in
Silver Spring, Maryland, contain the records of the AFL beginning in 1881
and more complete records of the AFL-CIO from the merger of the two
organizations in 1955 forward. The Walter P. Reuther Library at Wayne
State University has the largest labor collection, including the national
and many state and local records of the American Federation of State,
County, and Municipal Employees (AFSCME), the American Federation
of Teachers (AFT), the Congress of Industrial Organizations (CIO) prior
to the merger, the International Longshoremen, the International Workers
of the World (IWW), the Newspaper Guild, the Service Employees Inter-
national Union (SEIU), the United Auto Workers (UAW), the United Farm
Workers (UFW), and labor-affiliated organizations such as the Committee
for National Health Insurance. The Reuther Library also contains the
papers of numerous national, state, and local labor leaders, as well as the
papers of the founder of the National Association for the Advancement
of Colored People (NAACP) and influential civil rights activists such as
Rosa Parks. The Southern Labor Archives at Georgia State University
contain the records of many of the local, state, and regional labor orga-
nizations that were active in the South as well as the national offices
of the International Association of Machinists (IAM), the Professional
Air Traffic Controllers Organization (PATCO), and the United Garment
Workers of America (UGWA). Other papers of labor leaders, activists,
and state- and local-level organizations are scattered across the United
States in various library collections. The Society of American Archivists
maintains a listing of labor archives that is accessible from its website.
Limited records of the AFL, the CIO, and the Amalgamated Clothing
Workers of America are available on microform from University Publi-
cations of America (UPA) through Lexis-Nexis. UPA also offers micro-
form records of some heavily studied groups such as the NAACP, the
Congress of Racial Equality (CORE), and the Southern Christian Leader-
ship Conference (SCLC), as well as records from groups associated with
the Black Power Movement.
The availability of records for other types of organizations is more vari-

able. Groups’ archives can be located in a range of places, from private
libraries to government document repositories. The Hagley Library near
Wilmington, Delaware, holds collections from the National Association
of Manufacturers (NAM) and the Chamber of Commerce as well as
other trade associations and corporate offices. The Society of American
Archivists also maintains a listing of corporate archives, which can be
found on its website. The Library of Congress holds the records of civil
rights organizations such as the National Urban League and the Leader-
ship Conference on Civil Rights, public interest organizations such as the
National Consumers’ League, women’s rights organizations such as the
League of Women Voters, and materials from organizations fighting for
and against the Equal Rights Amendment. Limited records of the National
Organization of Women (NOW) and the National Abortion Rights Action
League (NARAL) are available at the Arthur and Elizabeth Schlesinger
Library on the History of Women in America at Harvard University.
Archives of the American Civil Liberties Union (ACLU) are located at the
Princeton University Library and the ACLU’s national office. The Sierra
Club’s records are held by the Bancroft Library of the University of Cali-
fornia, Berkeley. Though most archives are open to academic researchers,
some may be very difficult to access, such as those of the American
Medical Association, which are available only to association members.
To determine the availability of records for organizations that may be of
interest in a study, a researcher should first find what is available on the
Internet and contact the organization if it is still in existence.
The degree of completeness of records varies considerably across group

archives, but a range of materials may be available. Internal documents,
memoranda, and correspondence with politicians and policy makers, other
group leaders, and group members can be particularly helpful in under-
standing a group’s policy positions, strategies, and effectiveness. Group
archives also often contain other useful materials, such as journals and
newspapers published by the organizations and convention proceedings
that reveal a lot about an organization’s policy priorities. When they
are available, the minutes of leadership meetings can also be used to
gain information about internal deliberations and decision making. Some
archives contain press releases and pamphlets directed to members or the
general public that reveal information about policy positions and polit-
ical strategies. Some archives will also include figures on membership
and budgets that suggest the strength, stability, or insecurity of the orga-
nization. A few archives have collected transcripts of oral history inter-
views with prominent figures in the group or political allies; these may
be subjective, but they are still likely to enrich the understanding of the
group’s activities.
Unfortunately, the records of many organizations have not been

collected or are incomplete, but there are alternative prospects for finding
information on groups’ political and policy influence. A researcher should
identify important leaders, officers, and staff members and try to find out
if these individuals have contributed their papers to an archive. Despite the
fact that the AMA does not allow researchers access to its records, Chapin
was able to gain significant information on the organization’s activities
through the papers of one of its executive vice presidents and various
proceedings published in the Journal of the American Medical Associa-
tion (JAMA). A researcher should also contact the organization to see if it
maintains records that have not been processed that he or she might gain
access to. For example, Common Cause staff gave McFarland access to
staff memos and polling data, Tichenor gained access to documents from
the National Council of La Raza (NCLR) through its Washington office,
and the staff of the National Small Business Association (NSBA) allowed
Young to peruse unprocessed records housed at its main office. Moreover,
even if the records of the national office of an organization have not been
preserved, the records of state and local chapters may have been collected
by local libraries. These resources made possible Clemens’s and Skocpol’s
research into the activities of local grassroots organizations in the nine-
teenth and early twentieth centuries, which revealed a high level of group
influence that was not as apparent on the national level.
As illustrated in many of the examples of archival research discussed

earlier, a researcher will also likely find considerable information on
groups’ political activities and the development of policies of interest
to the groups that are being studied at the archives of federal agencies
collected by the National Archives and Records Administration (NARA),
presidential papers, congressional records, and papers from other polit-
ical institutions. For example, both Hacker and Klein were able to use
records of the Social Security Administration to gain more information on
the positions of labor and business groups on health and pension issues.
Martin used presidential records to look at the origins of the NAM, and
Dark and I used presidential records to evaluate labor’s political influence.
Research into the papers of members of Congress may also prove fruitful,
as it did with Chen’s study of the development of affirmative action policy.
The Society of American Archivists has even recommended that archivists
maintaining the records of members of Congress consider acquiring the
records of important groups associated with that member’s career (Phillips
1995). The papers of party leaders and party records located in presiden-
tial archives could also be useful in exploring interest group influence. A
range of government archives may even be available at the local level. For
example, in her study of the motivations of black women environmental
justice advocates in Memphis, Andrea Simpson was able to incorporate
research in the minutes of public meetings and records of the local archives
of the Defense Logistics Agency, a division of the Department of Defense
(Simpson, 2011). With a little legwork, archival research can likely be
incorporated into most research on the political activities of influential
groups.
Although the organization of records will vary considerably across
archives, there are several things that can be done in advance before
visiting any of them. Most of the archives maintain online finding aids, and
if they are not available, some archives may be willing to send or e-mail
copies of the aids for the topics that are most important to one’s research. It
is a good idea to review these in advance as well as to contact an archivist
to get a sense of how much material relevant to one’s topics of interest
the archives may hold. This will help one judge whether to hire a local
researcher or to visit oneself and for how long. In order to navigate the
finding aids, a researcher should get as much information as possible about
the organization and its activities. Figure out which leaders and staffers
in the organization, congressional office, or administration deal with the
issues of interest. Know which divisions of the organization are respon-
sible for what. It is also helpful to develop a time line of relevant dates,
such as when legislation was proposed or passed or when important deci-
sions were made. The secondary literature as well as contemporary press
accounts can be useful for putting together this information. For example,
I found the CQ Almanac very helpful for finding background informa-
tion on legislative battles because it covers every stage in the legislative
process and often provides fairly detailed information on interest groups’
activity on an issue. In addition to doing this type of preparatory work, it
is a good idea to explore any documents or oral histories that the archives
may have available online in advance to see if they suggest avenues for
further research.
A researcher will need to develop a system of keeping track of the

research. Before going, find out the archives’ policies regarding computer
use, note taking, and photocopying. While there, keep a list of all the boxes
and folders that have been looked at. Look at the archives’ recommended
form of citation and make sure to catalog all of the information that is
needed to properly cite a document. Some archives allow researchers to
make photocopies, whereas others require that photocopies be made by
their own staff. It is a good idea to write the citation information on the
photocopy if the archivist does not attach it. Most archives will now allow
researchers to take digital pictures of documents, which can really save
time and money. Before going, snap some test pictures of paperwork
and print them to figure out which settings work best on the camera. A
researcher will also need to be very careful about keeping track of where
the documents that are photographed are located. I wrote down (or typed
up) the picture number along with the subject and the citation information
in a log. At night, I checked back over my logs for accuracy and checked
through the pictures for the clarity of the images, copied them onto my
hard drive, and backed them up to alternative media. After spending hours
in my basement trying to locate a log notebook, I also started making
PDFs of these logs so I would not have to worry about losing them. It
might be helpful to snap pictures of box and folder titles as the research
proceeds, which can also help to keep track of where the documents are
located. Another benefit of taking digital pictures is that a researcher can
minimize the time spent at the archive by snapping a shot of anything that
looks interesting and reading it later. Of course, this means that some new
directions for research may be missed while a researcher is at the archive.
But if several days are being spent in the research, some time can be taken
at night to read back over the documents that were collected that day.
Future�Research
Because so few researchers have used archives to study interest groups,
there are a wealth of possibilities for future research. The archives of
members of Congress have been particularly underutilized. Because of

pressures of time, family, and the tenure clock, I relied on the archives of
labor organizations and presidential administrations to collect evidence of
labor’s influence in Congress. But I am sure my study would have been
greatly enriched by research in congressional archives. To date, most of
what is known about the relationship between interest groups and congres-
sional leaders has been gleaned through press accounts and interviews.
Congressional archives have barely been tapped to investigate impor-
tant questions regarding groups’ influence over leadership decisions and
how leaders employ group resources to deliver the votes of members of
Congress. Although some research has already been done in this area,
there is more room for the use of group archives in the study of the internal
dynamics of groups, such as how they attract, hold, and mobilize members
and how they build coalitions and alliances with other groups. There also
has not been much archival research on the electoral strategies of interest
groups. A range of research agendas might benefit from a greater use of
archival resources.
Endnotes
1. “Chronology,” prepared by Andrew Biemiller, the AFL-CIO’s chief

lobbyist, for AFL-CIO president George Meany. Legislative Reference
Files, box 48, folder 18, George Meany Memorial Archives.
2. Ibid.
3. Transcript, Lawrence F. O’Brien Oral History Interview XIII, September
10, 1986, by Michael L. Gillette, Internet Copy, Lyndon Baines Johnson
Library, 13.
4. November 19, 1947, memo from Clark Clifford to Harry S. Truman. The
Clifford papers are available at http://www.trumanlibrary.org.
Bibliography
Battista, Andrew. 2008. The Revival of Labor Liberalism. Urbana: University

of Illinois Press.
Baumgartner, Frank R., Jeffrey M. Berry, Marie Hojnacki, David C. Kimball,
and Beth L. Leech. 2009. Lobbying and Policy Change: Who Wins, Who
Loses, and Why. Chicago: University of Chicago Press.
Berry, Jeffrey M. 1999. The New Liberalism: The Rising Power of Citizen
Groups. Washington, DC: Brookings Institution Press.
Chapin, Christy Ford. 2010. “The American Medical Association, Health
Insurance Association of America, and Creation of the Corporate Health
Care System.” Studies in American Political Development 24(2): 143–
167.
Chen, Anthony S. 2009. The Fifth Freedom: Jobs, Politics, and Civil Rights in
the United States, 1941–1972. Princeton, NJ: Princeton University Press.
Clemens, Elisabeth Stephanie. 1997. The People’s Lobby: Organizational
Innovation and the Rise of Interest Group Politics in the United States,
1890–1925. Chicago: University of Chicago Press.
Dark, Taylor E. 1999. The Unions and the Democrats: An Enduring Alliance.
Ithaca, NY: ILR Press.
Draper, Alan. 1994. Conflict of Interests: Organized Labor and the Civil
Rights Movement in the South, 1954–1968. Vol. 29. Ithaca, NY: ILR Press.
Fine, Janice, and Daniel J. Tichenor. 2009. “A Movement Wrestling: Amer-
ican Labor’s Enduring Struggle with Immigration, 1866–2007.” Studies
in American Political Development 23(1): 84–113.
Francia, Peter L. 2006. The Future of Organized Labor in American Politics.
New York: Columbia University Press.
Frymer, Paul. 2008. Black and Blue: African Americans, the Labor Move-
ment, and the Decline of the Democratic Party. Princeton, NJ: Princeton
University Press.
Goldstein, Kenneth M. 1999. Interest Groups, Lobbying, and Participation in
America. New York: Cambridge University Press.
Gottschalk, Marie. 2000. The Shadow Welfare State: Labor, Business, and the
Politics of Health Care in the United States. Ithaca, NY: ILR Press.
Gray, Virginia, and David Lowery. 1996. The Population Ecology of Interest
Representation: Lobbying Communities in the American States. Ann
Arbor: University of Michigan Press.
Greenstone, J. David. 1977. Labor in American Politics. Phoenix ed. Chicago:
University of Chicago Press.
Hacker, Jacob S. 2002. The Divided Welfare State: The Battle Over Public
and Private Social Benefits in the United States. New York: Cambridge
University Press.
Hansen, John Mark. 1991. Gaining Access: Congress and the Farm Lobby,
1919–1981. Chicago: University of Chicago Press.
Hattam, Victoria Charlotte. 1993. Labor Visions and State Power: The Origins
of Business Unionism in the United States. Princeton, NJ: Princeton
University Press.
Heinz, John P., Edward O. Laumann, Robert L. Nelson, and Robert H. Salis-
bury. 1993. The Hollow Core: Private Interests in National Policy Making.
Cambridge, MA: Harvard University Press.
Klein, Jennifer. 2003. For All These Rights: Business, Labor, and the Shaping
of America's Public-Private Welfare State. Princeton, NJ: Princeton
University Press.
Knoke, David. 1990. Organizing for Collective Action: The Political
Economies of Associations. New York: A. de Gruyter.
Kollman, Ken. 1998. Outside Lobbying: Public Opinion and Interest Group
Strategies. Princeton, NJ: Princeton University Press.
Martin, Cathie J. 2000. Stuck in Neutral: Business and the Politics of Human
Capital Investment Policy. Princeton, NJ: Princeton University Press.
Martin, Cathie Jo. 2006. “Sectional Parties, Divided Business.” Studies in
American Political Development 20(2): 160–184.
McFarland, Andrew S. 1984. Common Cause: Lobbying in the Public Interest.
Chatham, NJ: Chatham House Publishers.
Nownes, Anthony J. 2006. Total Lobbying: What Lobbyists Want (And How
They Try to Get It). New York: Cambridge University Press.
Phillips, Faye. 1995. “Congressional Papers: Collection Development Poli-

cies.” American Archivist 58(3): 258–269.
Roof, Tracy. 2011. American Labor, Congress, and the Welfare State, 1935–
2010. Baltimore, MD: Johns Hopkins University Press.
Schlozman, Kay Lehman, and John T. Tierney. 1986. Organized Interests and
American Democracy. New York: Harper & Row.
Simpson, Andrea. 2011. “From Stigma to Strategy: Calculated Identity Posi-
tion and Black Women Activists.” Paper presented at the annual meeting
of the American Political Science Association, Seattle, WA.
Skocpol, Theda. 1992. Protecting Soldiers and Mothers: The Political Origins
of Social Policy in the United States. Cambridge, MA: Belknap Press of
Harvard University Press.
Tichenor, Daniel J. 2002. Dividing Lines: The Politics of Immigration Control
in America. Princeton, NJ: Princeton University Press.
Tichenor, Daniel J., and Richard A. Harris. 2002. “Organized Interests and
American Political Development.” Political Science Quarterly 117(4):
587–612.
Walker, Jack L. Jr. 1991. Mobilizing Interest Groups in America: Patrons,
Professions, and Social Movements. Ann Arbor: University of Michigan
Press.
Young, McGee. 2010. Developing Interests: Organizational Change and the
Politics of Advocacy. Lawrence: University Press of Kansas.
Part�III
On�the�Road
Chapter 11
Hitting�the�Road�Without
Hitting�the�Potholes
Sean Q Kelly and Linda A. Whitaker
“I just said we’d make it across. I didn’t say anything about the
wheels staying on.”
—Rubin Carver in Road Trip
Archival research is an adventure—a mixture of anticipation and incon-

venience, but only rarely a total disaster. Whether by road trip or discount
airfare, the key to traveling well is having an openness to the experience
and an understanding of where one is headed. The prospect of discovering
new material, new data, and new knowledge should offset the typical frus-
trations of travel. When things go badly in an archive, it is often due to a
lack of communication and planning.1 This is not a casual enterprise. Lost
time and squandered opportunities can haunt a researcher long after the
return home. A certain readiness and a willingness to communicate will

go a long way in fending off complete research failures.
This chapter is a collaboration between a political scientist and an
archivist who want researchers to be successful. The intention is to draw
on our experiences from both sides of the research process to help a
researcher prepare for that first immersion into the archives. The end result
will be that a researcher’s best efforts go directly towards the research,
not the travel details. Proof of success will be a desire to do it again and
again. At the end of the book is an appendix with two worksheets to aid
in planning.
Battle�Plans
One’s previous research habits may not be a good preparation for working
in a new research environment far from home and without the built-
in safety nets. Travel, recording and digesting new information, and
adjusting to long hours of intense focus require a strategy that makes the
most of each visit. To succeed, one must communicate like a David Parker,
plan like a Sean Kelly or a Scott Frisch, prioritize like a Brandon Rotting-
haus, and accommodate like a Kathleen Sullivan or Patricia Strach. Fortu-
nately, these scholars have mapped the territory so that others, including
archivists and librarians, may follow.
Many scholars who are new to using primary sources and visiting
archives have never consulted a librarian, let alone an archivist—for
anything. This is a well-documented pattern. Academics accustomed to
the relatively solitary nature of research generally are reluctant to reach
out for assistance. Some researchers may not relish sharing the trajectory
of their research for fear of being scooped. For these and other reasons,
researchers often are not prepared for the communication that is required
to remain pointed in the right direction on the research road. If only
because money and time are in short supply, we offer this advice: Get over
it. The name of the archival research game is establishing relationships.
Hitting the Road Without Hitting the Potholes 333
One must engage the archivist, conveying one’s passion and the impor-
tance of one’s research—even if a certain problem has not yet been settled
upon. Enthusiasm is contagious. An intellectually engaged archivist is
one who is willing to go the distance with a researcher. The fact that
a researcher has committed the resources to travel and conduct archival
research indicates that he or she is well positioned to take the next step.
Contact the repository well in advance of a research visit—multiple

times, if necessary. Know the rules of engagement. Observe the usual
courtesies and conventions. Get off to a good start. Do not be coy
about the topic of research. Outline as much as is known at the current
stage. Describe the research and what is being sought in the collection.
Archivists are in the information business. Most archivists like this kind of
dialogue. Give examples of citations that seem promising or were recom-
mended by others. If there is an example of what kinds of documents are
of interest, scan it and send it to the archivist. A picture is worth a thou-
sand words. Upon seeing the document, the archivist may immediately
recall similar material. These suggestions will help the archivist get on the
right track.
Archivists may not necessarily think like political scientists, especially

quantitative-minded political scientists. Campaign polls or demographic
information related to redistricting may be immediately recognizable data,
but qualitative material may not be conceived of as data by an archivist.
For instance, the committee assignment request letters used by Frisch and
Kelly (2006) appear as qualitative data (letters) but were rendered by them
as quantitative data. That is an easy leap for a political scientist but may
not be so easy for an archivist. Keep this in mind: Be ready to demon-
strate how such ubiquitous material as issue mail, committee memos, and
personal correspondence can be transformed into usable data. Kelly and
Frisch’s experience is a good example. In their work on voting in the
Republican Committee on Committees, the data appeared as “scraps of
paper” to the archivist. At one point, as Kelly coded the votes over three
days, the archivist turned to him and said, “I always wondered if those
were of any use.” For the first time, researchers were able to completely
pry open the black box of the committee assignment process. Thinking in
terms of data sets or parts of collections that could be rendered quantita-
tively requires a recalibration of an archivist’s knowledge of collections.
Archivists will get there once they see what works and what does not for
a given project.
Ask to speak with the archivist or librarian who is most familiar with
the collection (or collections) that is relevant to the research subject. In the
best case, this is the person who processed the collection. The person who
processes the collection has a bird’s-eye view of the material and is thus
intimately familiar with it. However, as is often the case, collections may
be processed by student interns or temporary archivists. To compound the
“knowledge of collections” problem, reading rooms are often staffed by
paraprofessionals or students who may know little about the material a
researcher wants (and may care less). One should make sure that one is
talking to the person who is best able to help with the research.
Providing reference is not only an art and a science, it is a partnership. It

is useful for an archivist to know if a researcher is new to archival research
and about search strategies that were dead ends. For an archivist, knowing
what not to look for is as helpful as knowing what is sought. A scholar
has license to roam. Feel free to ask questions. How was the material
put together? Are the labels and descriptions reliable? Was the material
deposited serially or as a whole? Find out about the background of the
collection, how it came to be, and what gaps may exist. None of these
confounding factors may be reflected in the finding aid, particularly if the
collection was processed more than fifteen years before.
Archivists and librarians are in the business of making connections

among collections, other repositories, and other people. Is the donor still
alive? Would the donor be willing to speak with a researcher? Is the collec-
tion split or shared with another repository? What other repositories in the
area hold similar collections? Ask about material that may not be donated
anywhere but may be relevant to the project. Archivists can open up a
whole world beyond archives. They may be able to put a researcher in

touch with key campaign and office staff or family members; help locate
local government and business records; facilitate access to collections held
by nonprofits, churches, and private individuals; or connect a researcher
to other scholars who are conducting similar research.
Researcher-archivist relationships may span one’s entire research

career. For instance, our friendship goes back a number of years. Kelly
contacted Whitaker, who was processing the papers of Senator Dennis
DeConcini, to inquire about the existence of Senate committee request
letters that might be in DeConcini’s papers. Whitaker suspended her
processing to locate a potential box in the sprawling collection and
processed the letter and other material to make them available within
weeks.2
Even if the archivist does not offer, ask for an exit interview to discuss
how the visit went and what directions future visits may take. More than
likely, the archivist will inquire about the progress of the research more
than once. The interest is sincere, and archivists will listen carefully as a
researcher recounts his or her findings. Remember, a tremendous amount
of labor (salaries) went into arranging these collections, and significant
resources (space, shelving, climate control) are continually invested in
them. Archivists are gratified to learn about researchers’ successes. Many
also report to their boards and administrators the quantitative outcomes
(i.e., the number of scholarly publications) resulting from the research that
is done in the collections.3 So, stay in touch. The reward will be unex-
pected dividends, such as notices that the repository has something new,
reports of a lead that should be followed, or a key document that was found
after a researcher’s visit and is now sent as an e-mail attachment.
Making�Travel�Plans
The most important decision to be made is how many days to invest in the
trip. Investing too many days means wasting time and money. If one is
married, partnered, and/or has a family, the time is at least as important as

the money. Failing to budget enough time causes frustration and can actu-
ally cost more money (the costs of a return trip). Experience helps, but on
a first trip—or a second or third—the researcher’s best ally is the archivist.
Get an estimate of the amount of material and ask for the archivist’s esti-
mate of how many days might be needed to get through them. Be sure
to ask about the size of the boxes. Standard file boxes, which can hold
thousands of pages, are substantially larger than archival document boxes,
which hold seven hundred to nine hundred pages. If one is told there are
ten boxes, for example, knowing the type of box is an important detail. A
researcher may also be offered access to unprocessed material. If that is
the case, navigating those boxes could easily take twice the time.
Before going, do some research into the other collections at the reposi-
tory as well as other nearby locations. This is useful for two reasons: (1)
collections may be found in the area that bear on the topic, and (2) if extra
time is available, it may be worthwhile to poke into these collections to
see if there is useful material.
Whether a researcher has university or repository funding or is person-

ally underwriting the trip, the research is probably being done on a shoe-
string. Budgeting is important. In the sections that follow, we try to
consider all of the costs a researcher might face.
Travel
Carefully consider the most economical way to get to the destination. If
the distance is significant, flying is probably the only option. However,
look carefully at whether driving to closer locations makes sense. First
and most important, does the funding source allow reimbursement for
mileage? Many do not. If the funding source does not cover mileage, a
researcher will be on the hook for the gas and wear and tear on his or her
car. Some destinations are expensive to fly to (smaller towns, for instance);
it may be more economical to fly to a nearby airport and take a train or
bus or even rent a car and drive (which adds mobility at the location, but
do not forget the costs and logistics of parking on campus!).
Do not forget to factor in the costs of ground transportation. It will be
necessary to get back and forth to the originating airport or train station
and to pay for parking; it will also be necessary to get back and forth to
the station at the destination. If it is not possible to stay within walking
distance of the repository, figure those ground transportation costs in, too.
Lodging
Many repositories list lodging and other visitor information. This informa-
tion, however, is not always up to date. Note that “closest” and “cheapest”
could mean thin walls and a lumpy bed—the cost savings may not be
worth several nights of poor sleep. Check with archival staff to get
“insider” information about where to stay. Remember, they live there; they
will know how far hotels are from their repositories and may have some
insight about the quality of accommodations. Archivists understand trav-
eling on a tight budget and tight schedules. Because research may take one
to communities far from cities, do not hesitate to contact the repository.
Generally, the staff will be friendly and eager to provide information that
may save time and money. Some institutions may provide on-site, inex-
pensive rates, but one needs to ask. During the summer, many campuses
rent out dormitory rooms to researchers; this can be an economical and
convenient option (though we are not sure we recommend the food!).
Repository�Fees
Factor in repository fees at the front end when planning a budget and/or
applying for funding, regardless of the source. Photocopying and image
reproduction fees vary considerably among repositories. These are usually
listed on the archives website. Do not expect to find bulk discounts.4 A
good rule of thumb is that if the material looks pertinent or the image fits,
get it then and there. Chasing it long-distance will be expensive; returning

to get it will be even more expensive. Archivists understand that time is
money. Reproductions take time and are rarely free. Let the reading room
staff know when the visit will end so that they can plan accordingly.
Miscellaneous�Equipment�Costs
Some other costs are worth taking into consideration. These are one-time
costs for equipment and materials that will be used repeatedly. A laptop
is useful for many applications. If a digital camera will be used (where
allowed), procure a good one (10 megapixels minimum) with multiple
rechargeable batteries and plenty of secure digital storage. A spare hard
drive will also likely come in handy. Each image from a 10-MP camera
takes up about 4 MB of hard drive space; a laptop hard drive will be filled
up quickly at that pace. The external hard drive is a convenient place to
back up one’s hard (and expensive) work collecting all of those images.
A tripod that allows the camera to be pointed down at a 90-degree angle
is also useful; holding a camera while taking hundreds or thousands of
photos can be painful.
Searching�for�Funding
By now, the mantra should be clear: The limiting factors in archival
research are time and money. There is never enough of either. Perhaps the
most practical concern of congressional scholars and other scholars inter-
ested in archival research is money. Many congressional archives have
funding available to support travel to the archive and related expenses.5
Usually, funding is contingent on the researcher having a clear idea about
what he or she is looking for, explained in a short essay, and whether
the holdings in the archive contain information relevant to the research
project. This is where a strong research design helps. It is important to be
able to explain what one is looking for so that the archive has a better idea
about if and how its resources can benefit a scholar’s research.
If the archive has a website—many do, and the number is growing

rapidly—check for finding aids that provide a somewhat detailed index
to the contents of the collection to determine if it has materials that can
aid in the research. Small repositories may not have the funds or technical
infrastructure to post finding aids on the Internet. Most will charge for the
duplication of and postage to mail hard copies. They may not be able to
take credit cards. Be prepared to send a check and wait for delivery by
postal mail. A conversation with the archivist will help narrow the pages
that are needed. If there is a fixed dollar amount available to spend, that
is not a problem. Repositories will copy up to that amount upon request.6
Do not hesitate to contact an archivist at the repository and explain

the project. This has two benefits. First, one can get a better idea about
whether the archive will be useful to one’s research. Second, making
contact with an archive that may be funding one’s trip will give one’s
grant application a higher profile in the selection process. It may help
to “sample” a promising folder or two by requesting a few documents
before committing to a visit. For a fee, archivists may photocopy or scan
a few pages for review. Be sure to mention in a proposal that contact has
been made and that the archive has the relevant data. Appropriate back-
ground preparation, combined with the fact that very few scholars (espe-
cially political scientists) use these archives, means that an application has
a very good chance of being approved. These stipends vary from a few
hundred dollars to more than a thousand dollars.
Sometimes the costs of travel and document duplication (which varies

from ten to fifty cents per page) and postage exceed the available stipend.7
In these cases, internal university funding is perhaps the most conve-
nient source of funding. Most universities have small grant programs for
faculty; larger graduate institutions offer small amounts of funding for
graduate student research. For graduate students, these small grant awards
offer an excellent opportunity to simultaneously hone grant-writing skills
and build a vita. For early-career professors, such grant awards are a good
way to demonstrate immediate success to the department and administra-
tion while waiting for the slow wheels of the journal machine to begin to
provide more evidence of research accomplishments.
Consider some other creative ways of underwriting archival trips—for
instance, combining research trips with trips to professional conferences,
for which many universities offer funding. While in town for the confer-
ence, make use of a local collection or rent a car to drive to a nearby repos-
itory. Some researchers have been known to combine family visits or parts
of vacations with detours to useful collections.
Boots�on�the�Ground
So much for the warnings and admonitions. What does the actual
experience feel like? At first, archival research is intimidating. It is
nerve-wracking. It is full of promise and the possibility of being less
than successful. Archival research is also intellectually stimulating and
exciting; all of the authors included in this volume agree on this point.
Here is another mantra: Do the homework. Arrive prepared for the initial
foray. We have said it before and we will say it again. One measure of
success in archival research is finding the material one needs. The real
proof of success, however, is whether one makes second, third, and fourth
trips to archives.
Expectations
Archival research is more physically and mentally demanding than might
be expected. Sitting and/or standing in a reading room for eight hours puts
strain on the body. Reading (or at least skimming) hundreds or thousands
of pages per day, mentally straining to determine the logical structure of
the papers and to extract the meaning of the huge volume of the mate-
rial that is embedded in the broader context of the collection, is mentally
fatiguing. Think about it: a researcher is trying to make sense of written
material that, in real time, developed over the course of months and years.
It is difficult work that requires stamina and focus.
Whether it is a first visit to an archive or a one hundredth visit, keep
personal expectations low. Be prepared for the differences between the
folder title that can be seen online and what is found in the reading room.8
The title will give no indication regarding bulk (or lack thereof) or the type
of print material that is included, and it may not even reflect the actual
content. All researchers want to find that one document or that one folder
that proves their case. Let us be realistic: despite what is shown in the
movies or on television, there are no smoking guns; there is no single
source that will allow one to crack the “da Vinci code.” Sometimes there
will be material that is very useful, even revelatory, but it is also possible
that there will be less than might be hoped. If a folder is particularly disap-
pointing, there may be reasons for it, especially if the original folder labels
have been retained. (What made sense to the original records filer may
make no sense to the researcher or the archivist.) Further, there may be no
consistency for how certain materials were filed. If frustration sets in, ask
the archivists for assistance. They are pattern-recognition experts. Mean-
while, keeping expectations low ensures that they will almost always be
exceeded.
Request material commensurate with the time that is available. Archival

research cannot be done hastily or sandwiched between other appoint-
ments. In the beginning, the first few boxes will likely take longer to work
through than one has estimated. Early on, one is trying to wrap one’s head
around the structure of sets of records that were created by others and
arranged by others; two boxes may take from two to four hours. Do not
worry. Speed will pick up as one develops a feel for the structure of the
collection and as document recognition improves.
Request the most promising boxes first. This may seem banal advice,
but there is a strong inclination to proceed through the collection chrono-
logically or by series. This tactic may not always be appropriate (although
sometimes it is). Let the research design, not the structure of the collec-
tion, dictate the approach.
Research is a process. It is fluid. Be flexible regarding the method-
ology. Be open to suggestions to look at material other than what was
originally requested. Personal collections, marketing research materials,
nonprofits, newspaper archives, quirky in-house databases, oral histories,
film, and personal interviews are a parallel universe that holds the promise
of untapped data. A researcher is limited only by his or her imagination
to use it.
Be sure to take breaks. Archival research is mentally and physically

taxing. Get up and stretch, walk around, and get a drink of water. Because
of limited time and the fascination one forms for the collection, one is
apt to postpone breaks or skip lunch. Some repositories close for a lunch
hour, forcing one to stop. If there is not an enforced lunch break, one
should impose it on oneself. Without breaks, one will likely develop
a headache, become cranky, and lose one’s mental edge. None of this
helps the research. It may seem like silly advice, but all of the authors
(and the archivists) in this volume endorse it. These are lessons learned
during hundreds of days and thousands of hours conducting archival
research. The contributors to this volume know firsthand that inadequate
rest, hydration, and caloric intake can take their toll.
At the end of each day, take a few minutes to jot down thoughts, ideas,
and reactions. These notes can help chart future research or identify poten-
tial publications or presentations in the near term. It may help a researcher
to communicate more clearly with his or her coauthor, mentor(s), or thesis
committee about the directions his or her research might take or how
archival materials might be used in the classroom to illustrate a point.
Security
Archivists take the stewardship of their collections seriously. A major
responsibility is to insure that collections remain intact and available to
researchers well into the future. Most archives will prohibit backpacks,
briefcases, and even personal papers, notebooks, and the like in reading
rooms. The vast majority of repositories allow laptop computers in the
reading room, but be prepared to open the laptop on entry and exit to prove
that no documents are present.
Most rules are security measures intended to protect against the theft or
damage to rare, unique, or fragile materials. The highly publicized case
of former national security advisor Sandy Berger, who tried to purloin
papers from the records of his work during the Clinton administration
from the National Archives, illustrates the most extreme concern for
archivists: theft. The more realistic concern, however, is that documents
will mistakenly find their way into a researcher’s personal materials and
disappear. The archival book stacks and storage areas are routinely closed
to researchers and are accessible only by staff.
Upon arriving, a visitor will be asked to complete a researcher applica-

tion, provide local contact information, and provide identification. If it is
an option, fill out the paperwork and send it to the repository before arrival.
This will minimize the initial “administrivia” at the beginning of the visit,
creating more time to spend talking with the archivist and getting into the
papers. There may be any number of additional precautions imposed or
forms to sign that indicate the rules have been read and understood. Do
not fight it; just do it.
Security procedures at the National Archives are the most extensive one
will find (in part because of the Sandy Berger incident and in part because
of post–September 11 security concerns, but who knows what the National
Treasure movie series has added to that general concern about security!).
Be prepared to invest at least an hour during the first trip to complete the
researcher application process and self-paced NARA researcher training.
Bring only what is absolutely necessary. Each piece of electronic equip-

ment will be documented and any papers a researcher has when he or she
leaves will be searched page by page.
Handling�the�Materials
Each reading room will have rules about how documents should be
handled. Typically, the archive will limit the number of boxes researchers
can request at any given time. Most will require that researchers work
with one box at a time and that only one folder at a time be removed from
a box. Keep documents flat and on the table. Keep them away from the
edges of the table to avoid the embarrassing and potentially destructive
possibility that the papers spill onto the floor. No repository with which
we are familiar allows pens in the reading room—pencils only. Gloves
are required for handling photographs, but some repositories will insist
researchers use gloves for all print materials. If a researcher is rough, care-
less, or does not heed warnings, he or she may be asked to leave or barred
from returning. This is rare, but it does happen.
Reproduction
Most archives allow some form of reproduction of their materials for
use by researchers. Photocopying is typically allowed. Some repositories
allow researchers to use equipment unsupervised. If this is the case, deter-
mine the repository’s guidelines for handling materials that are stapled,
paper clipped, binder clipped, or oversized. Staff are protective of their
materials, and each repository has different ways of dealing with these
issues. For instance, at the Reagan Presidential Library, the staff insists on
removing staples and clipping documents together; the Carter Presidential
Library does not follow this convention. When doing self-service copying,
carefully note the location (collection, series, box, and folder numbers)
on the copies as they are made. This step is vital for citing documents in
research and critical if a researcher needs to return to a particular point in
the collection. Asking archivists to retrace one’s steps after the fact will
not win one friends.9
Other repositories require that staff copy materials. There are certain
advantages to this. First, time not spent at a copier is time one can spend
working with the papers; copying can be dull work if hundreds or thou-
sands of pages are involved. Second, standing at a copier can induce some
of the most mind-numbing backaches one will ever experience. Third,
repository staffs typically do an outstanding job of documenting the source
of each document. If staff will copy the materials, provide very clear guid-
ance about which pages should be copied. There is nothing worse than
doing all of that hard work and then getting home to find that a critical
document is missing.
Up until recently, photocopying, at the rate imposed by the archive,

was the only option for researchers. Some repositories now provide scan-
ners for the use of researchers or will scan and send documents via e-
mail. Many repositories (NARA included) allow researchers to use digital
cameras. Ask first and be familiar with the repository’s specific policy.
Most will not allow the use of flash, so ensure that an adequate light source
is available. Documenting the source of material when using a digital
camera is particularly important. Remember to take a picture of each box
and folder a document comes out of; “tagging” photos using photo orga-
nizing software—such as Picasa™, which is free—will help one locate
photos quickly.
Keep in mind that having a copy of a document does not make one the
“owner” of the intellectual material. Classroom and thesis use meets the
Fair Use Act10 criteria, but be careful how the material is reproduced and
distributed, especially when posting it to the Internet. There is nothing a
university counsel (or personal legal counsel) will appreciate more than
keeping the institution out of an intellectual property rights lawsuit.
Taking�Notes�or�Making�Copies
Efficiencies are in the eye of the beholder. Taking notes may save copying
costs, but it takes more time. If one is a note taker, one should take more
notes than what one deems necessary. When one is engaged with the
papers, one often does not have a complete idea of what material will end
up being used or how the notes will shape one’s thinking and the intellec-
tual development of the research. Taking broad notes can provide much
value later during analysis and writing. Identifying documents to copy
now and read later may save time, but it will not necessarily save money.
Choose wisely.
If a researcher is visiting multiple repositories and using multiple collec-

tions, citations become critical. It is easy to get lost on the research road.
This bears repeating: The biggest mistake new researchers make is failing
to note where the document is located. The finding aid front matter will
often state the preferred citation for the collection. Many researchers waste
valuable time retracing steps to confirm box and folder numbers.
Other�Considerations:�Hidden�Collections
If a scholar is planning archival research, then he or she is already pushing
the envelope within the discipline of political science. Once the basic
challenges and risks have been accepted, one may be prepared to go a
step further, especially after a successful series of first visits—into the
unprocessed collections that may or may not be found in catalogs or on
websites. Some seasoned archival researchers (e.g., historians) say it takes
twice as long to navigate unprocessed collections. They will use them
as a last resort if it is critical to their project. However, getting access
to unprocessed collections can result in new information that leads to
a journal article, an invitation to present a paper at a national confer-
ence, or a breakthrough in research. There is a trend nationwide to make
unprocessed collections available.
Citations for unprocessed, partially processed, or minimally processed

collections are tricky. The trade-off for accessing “hidden” or relatively
unknown material in the pursuit of fresh data is grounding that new infor-
mation with a reliable citation. Ask the archivist how to cite these collec-
tions because box and folder numbers may not be known, an official title
to the collection may not have been given, the physical location of the
collection may change, and only an acquisition number may exist.
Look beyond traditional print sources. Speeches, hearings, press confer-

ences, guest appearances, panel discussions, and town hall meetings may
be captured on radio and television tape. If this is of interest and the mate-
rial is noted in the finding aid, ask if it is possible to view it. If the repos-
itory does not have the equipment, ask if it can be transferred to a CD
or DVD. Most archives will have access to a vendor or may even do it
themselves. Note that researchers will often be asked to pay the costs of
digitizing these older formats.
Restrictions�and�Restricted�Collections
Most finding aids will state restrictions up front. Be sure to fully under-
stand these restrictions. Nothing would be more frustrating than to make
the effort to visit a collection and find a critical component closed for use.
Archivists try to limit restrictions. Most will not accept collections with
undue restrictions. If restrictions are allowed, there should be a specific
end date. Repositories usually have a policy for removing restrictions,
though there is no guarantee that access will be granted. Archivists have
been known to approach donors and other repositories to change deeds
of gift. If possible, try approaching the donor directly. Explain the nature
of the research and the materials that are of interest. Kelly has found
that former politicians are more wary of journalists—who may wind up
embarrassing the donor—than of academics with legitimate intellectual
interests (the reasoning perhaps being that nobody reads academic writing
anyway!). In Kelly’s experience, the vast majority of direct requests are
granted. An upshot of this approach is that the donor may take a legitimate
interest in a researcher’s work and grant an interview.
Federal agencies such as NARA and presidential libraries post infor-
mation about the Freedom of Information Act (FOIA) process. Here is
fair warning: This takes time. If a project depends on access, a researcher
will have to mount a very persuasive case. Otherwise, he or she will be at
the end of a lengthy queue. For instance, Kelly submitted a FOIA request
to the Clinton Presidential Library in 2007. As of this writing, he is still
waiting to gain access to the requested documents.
Distance�Research
It can be risky, but every researcher tries distance research at least once. It
works best if the staff knows the researcher and the project. On average,
four e-mail exchanges are required before items can be pulled. E-mailing
an example of what one is looking for (if there is one) is particularly useful.
Most repositories post a time limit (usually thirty minutes) to search for
the material a researcher wants before a research fee is charged. Then the
meter starts ticking. A premium is often placed on photocopying, shipping
and handling, and so on. According to Whitaker, researchers invariably
ask, “So, what’s in there? How much material is it?” The answer will be
a quick appraisal and guesstimate. Because a researcher is dealing with
a non–field expert, he or she runs the risk of running up a significant tab
and gathering quantities of superfluous material which may not yield the
desired information.
In some cases, it may be possible to hire a local researcher to help

with research. For their work on committee assignments, Frisch and Kelly
hired a history graduate student at Indiana University to work through
an unprocessed congressional collection to locate committee assignment
request letters. They contacted the chair of the department and asked him
to broadcast the opportunity to graduate students. Several responded, and
one was particularly enthusiastic. The authors used the history department
because political science graduate students are not trained for archival
research (hence this book); they may not even know that their university
has important congressional papers collections.
Other�Researchers
Ask if other researchers have used this material in the same way or if
they are exploring the same topic. There is synergy when researchers
are connected to one another. This requires a balance between promoting
collegial exchange and preserving privacy and confidentiality. Many
archivists will contact researchers and ask permission to use their contact
information for other researchers who follow.
If one knows that another researcher has used the collection for similar
purposes, one should share this information with the archivist. Reposito-
ries maintain a record of researchers and the boxes that they have accessed.
By returning to that record, the archivist may be able to go directly to
boxes that are of interest in this case. For their work on Senate committee
assignments, Frisch and Kelly discovered that letting the archivist know
about previous research in the Richard Russell papers resulted in the
archivist quickly locating the materials that they were looking for.
There is one last thing: archives and archival research can be just plain
fun. The scholar is on walkabout—off campus and away from home—free
to test his or her research mettle, make mistakes, and share discoveries.
Although it requires focus and discipline, there is a certain freedom to
exercising a brain unrestrained by lectures, secondary sources, and thesis
committee members. The added bonus is that archivists provide a ready
audience for researchers’ ideas, observations, and responses to the mate-
rials they have pulled. The feeling of pure joy and personal satisfaction
that comes with discovery may be rare and may be fleeting, but it is a
common event in reading rooms. Why deny oneself the experience?
Conclusions
A carefully planned trip is critical to success in archival research. Part of
being prepared is developing a strong research design. We assume that
our readers will be fine in that department. Like any successful military
campaign, a research strategy must be well planned and well executed.
From getting to the repository to “living” there and making the day as
productive as possible—that is where we hope this chapter has helped to
identify the fine details of the trip. Be sure to go over the particulars care-
fully. Do the homework (working through the finding aid, communicating
with the archivist, knowing the policies of the archive, and the like), and
the first trip will be a breeze, and subsequent trips will be productive too.
If one is fortunate enough to be located near a political papers collection,

try doing a “dry run” using an interesting, if not appropriate, collection.
At a large university, visit the special collections department to get a feel
for the ambience of the reading room. Request and look through a few
boxes. Although the protocol for calling boxes will not be the same at
every institution, they are usually similar. Talk with the archivist. Tell the
archivist what the intent of the visit is and let him or her know one is trying
to learn. Chances are he or she can provide some advice.
Archivists are not “dataheads” (like many political scientists), but they
can be educated and they can adapt. The stewards of congressional collec-
tions want to convince researchers of the value of archives. They believe
that archives, and these collections in particular, hold pure potential for
the scholar with an open mind. Political papers are underutilized. They
are voluminous, sometimes disorganized, and in some cases a challenge.
They are an acquired taste. But for those scholars who use them and for
those archivists who take care of them, these collections are a limitless
source of fascination and knowledge.
The prospect of political scientists flocking to archival research as new

users has the potential to revitalize both professions. Most archivists like
to teach. This offers real opportunities to guide scholars early in their
careers. (Veteran users of archives tend to resist suggestions and, based

on our observations, become more rigid over time.) The opportunity to
provide classroom instruction for graduate students in political science,
the chance to work collaboratively on books and presentations, and the
option of assisting with data collection is not only intellectually stimu-
lating, it also improves archival practice.11
To help with planning, use the budget and travel tools on the next couple
of pages. Good luck!
Endnotes
1. What are the worst-case scenarios? The repository is unable to locate the
collection, or the requested folders are missing. The repository is closed on
the day one arrives. The repository has succumbed to a natural or unnat-
ural disaster.
2. See chapter 2. What Whitaker and her colleague Rebecca Hankins did
was contact Senator DeConcini to see if he would change the deed of
gift to allow researchers access to his largely unprocessed collection. His
response was quick and positive, resulting in a modification that took less
than a month to complete. (Note that the original deed of gift took nearly
a year to craft and negotiate.) In the archives world, this is the equivalent
of moving heaven and earth. It is also proof positive that deeds of gift can
be changed, particularly if donor relations are good.
3. As a fine example, the Rockefeller Archives Center provides yearly
reports of scholarly output based on the use of its collections.
4. Fees are designed, among other things, to discourage mass reproductions
of archival material. Repositories are not in the business of populating
personal libraries or other archives with their collections.
5. The Albert Center, the Dirksen Center, and the Ford Presidential
Library all have travel grants. Several other collections also offer
funding, such as former representative Morris Udall’s (D-AZ) collec-
tion and former representative Claude Pepper’s (D-FL) collection, for
example. For a more complete listing, see the Society of American
Archivists Congressional Papers Roundtable site: http://www2.archivists.
org/groups/congressional-papers-roundtable/grants-and-fellowships-1.
6. Reproduction fees for print materials are not money-makers for repos-
itories. The fees typically cover the costs of paper, the maintenance of
copying machines, and in some cases, the rental fees of the equipment.
7. One of the authors confesses to avoiding postage costs by packing an
empty suitcase and filling it with photocopies. We recommend including
this suitcase as a part of carry-on luggage rather than risking having one’s
hard work misdirected to Boise (unless, of course, the researcher lives in
Boise!).
8. The differences between the virtual representation and the tangible item
cannot be overstated. See E. Yakel and D. Torres. 2003. “AI: Archival
Intelligence and User Expertise.” The American Archivist 66(1): 51–78.
9. Yes, it can and has been done, but do not expect the same enthusiasm and
interest for this task compared to the initial searching effort.
10. “Fair use” is poorly understood by scholars who are used to free and
unlimited access to information. To be published without consequences,
know the legal guidelines. Cornell’s Copyright Information Center (http://
www.copyright.cornell.edu/) offers an excellent tutorial, has a separate
section for faculty and staff, and provides a clearance services page for
users.
11. Rudy Espino, assistant professor of political science, Arizona State
University, has used the reading room at the Arizona Historical Founda-
tion as a “laboratory” for data collection, giving students hands-on expe-
rience with rare, fragile Arizona territorial documents. They tabulated and
plotted votes along ideological lines. These data were used to compare the
voting records of the territorial legislatures of Arizona and New Mexico.
Appendix�A
Table 10. Budget worksheet

Appendix�B
Table 11. Travel checklist

Index
Abraham, Terry, 121, 153 Appropriations Committee

Academy of Certified Archivists, (Senate), 51, 227, 231–232, 240
27 archival collections, businesses
AFL, 300–302, 310, 312, 317–318, 145; church archives 145; local
324 and state government, 145 275–
AFL-CIO, 300–302, 312, 317, 324 297; military 145; municipal
Agronsky and Company, 65 281–284; organizations, 89, 145,
Albert, Carl, 54, 62–67, 91, 95–96, 244, 252–254, 258 284; personal
272, 352 papers (general manuscript
American Civil Liberties Union collections), 88, 144, 213,
(ACLU), 318 243–244, 255–265
American exceptionalism, 83, 299 archival record: interpreting,
American Federation of State, 65–66, 82, 85, 90–92, 173–174,
County, and Municipal 176–177, 221–223, 246–247,
Employees (AFSCME), 317 263–265, 289–290
American Federation of Teachers Archival Research Catalog (ARC),
(AFT), 317 215
American Heritage Center, 18, 111 archival research: barriers 40–41,
American labor movement, 299 71, 76 190–194 ; challenges 5–
American Medical Association 10, 40–41, 71–72, 76, 190–194,
(AMA), 311, 319–320, 325 215, 219–221, 230–231 237
American National Election Studies 263–265 269; critics of 5–10,
(ANES), 13, 162–163, 181, 187, 16–25 224; concept building
260 11, 66, 163–164, 170, 175,
American political development 243–245 246–247; costs 8–10,
(APD), 26–27, 54, 67, 71–72, 40–41, 54, 93, 336–338 356 ;
77, 82–84, 86–89, 93, 96–100, data generation process 13–14,
272, 304, 306, 310, 325–327, 23–25, 63, 88 247–249; political
363 science 1–2, 27, 61–94 245–253;
American Society for the risks 6–7, 76, 191, 222–223, 348
Prevention of Cruelty to Animals 352; search strategy 64, 141–145
(ASPCA), 292 236–237 255–260, 282, 284–285
American Voter, The, 39 336; serendipity 7, 15–16, 243,
Appropriations Committee (House), 244–245, 262285 349; teaching
51, 227, 231–232, 240 12–13 93; (continued on next
page)
archival research: theory building ArchivesUSA, 284

6, 11, 66–67, 77, 78–81, 90, archivists, 3, 9, 16, 18, 22, 27,
185–187, 245–246 263; time 54, 57, 70–71, 95, 101–113,
commitment 7–10, 40–41, 76, 115–116, 122, 125, 127–128,
93, 192–193 236 289, 290, 331, 131–133, 136–137, 139,
336, 340–341 142–143, 145, 147–148,
archives, 2, 4–5, 7–9, 14, 16–17, 153–157, 172, 196, 215–216,
19, 21–22, 26–31, 33, 36, 219–221, 236–237, 255, 261,
41–42, 46, 51, 54, 56–58, 281, 295, 318, 320, 332–335,
64–65, 68–72, 76–77, 79–80, 337–339, 341–343, 345, 347,
82, 88–90, 92, 94–96, 99–107, 349–350, 352 363; working
111, 113, 116–117, 127–128, with, 95, 191–194, 216, 221
131, 136–137, 140, 143–151, 237 261, 286, 331–336 344–
153–156, 159, 161, 163–165, 347; responsibility of, 16–17,
167, 169–170, 172, 174–175, 52, 103, 343 347; role 3, 27, 70,
179–180, 182, 186–189, 191, 103–105, 331–335
194–196, 198, 200–201, 204, Arizona Historical Foundation, 123,
206–208, 210–214, 216–218, 137, 180, 196–199, 201–202,
220–221, 225–226, 228–233, 353, 363
235–238, 240, 243–247, Arizona State University Libraries,
249–251, 253, 255, 257–269, 123, 134–135, 156, 180,
271, 273, 275–276, 278, 197–198, 201–202, 240, 353,
281–286, 288–293, 295, 297, 361
300, 302–303, 307–324, 332, Armed Services Committee
335, 337–340, 342–344, 347, (House), 145, 227, 230, 240
349–352, 359–360, 363 Armed Services Committee
archives: appraisal, 2–3 113; (Senate), 145, 227, 230, 240
arrangement of, 3, 125, 206–215 Arnold, Doug, 40
261 ; backlog, 111–112 147; Arthur and Elizabeth Schlesinger
congressional, 4, 36, 37, Library on the History of
109–115, 119–128, 130–144, Women in America (Harvard
163, 165, 169–199 228 255–256; University), 318
locations, 4, 48, 49, 207, 229,
230, 232, 239254–260, 266, 268 Bachrach, Peter 174
281–284; manipulation, 20–22 Bancroft Library (University of
128; measurement, 5–6, 105, California, Berkeley), 319
178–184, 188–189 219–220; Baratz, Morton, 174
opportunities, 63, 67, 77, 79–80, Bates College, 139
85, 88–93, 175–190, 263–265 behavioral reality, 6, 63, 65, 67, 69,
286; provenance, 2–3 128; 71, 73, 75, 77, 79, 81, 83, 85, 87,
restrictions, 131–133, 191–192, 89, 91, 93–95, 97, 99
216–218, 347–348 behavioral revolution, 26, 37–39,
54, 266
Index 361
behavioralism, 68–70, 73, 76, 83, Center for Legislative Archives,

272 255, 257, 266, 268
Bensel, Richard, 71, 87, 96, 98 Central Intelligence Agency (CIA),
Berinsky, Adam, 254, 267, 270 227–235, 237–241, 359
Bevill, Tom, 42–43, 45–46 Chamber of Commerce,
Biemiller, Andrew, 300–303, 324 Church, Frank (senator,), 7, 103,
Bieneke Library (Yale University), 145, 164, 174, 186, 198, 200,
111 230, 268
Bingaman, Jeff (New Mexico church, 283–284, 291
attorney general), 150, 173 cities, 145, 275–280, 282–283,
Biographical Directory of the 285–286, 289, 292–293, 297,
United States Congress, 337, 363
1774–Present, 4, 229, 236, city records, 283–284, 291
255–256 Civic Club of Allegheny County,
Bridges, Styles, (senator), 229, 232, 284, 295
240 civil rights movement, 312–313,
bully pulpit, 203–204 325
business groups, 320 Clinton, William, 216, 221, 225,
299–300, 343, 348
Cahalan, Don, 253, 270 Cold War, 228–232, 235
campaign, 7, 13, 22, 24, 39, 47, collective bargaining, 299, 301, 310
63, 77, 99, 107, 109, 125, College of Idaho, 7
129–130, 134, 141, 144, 150, Columbia Center for Oral History,
162–166, 169–170, 172–174, 258, 266
176, 178–183, 185–191, 193, Columbia University Rare Books
195–196, 198, 200, 246–248, and Manuscripts Library, 266
253, 271, 295, 306, 333, 335, Committee Assignment Politics, 30,
350 32, 52, 57, 200, 360
campaign staffers, 174 committee assignments, 6, 13, 15,
campaign strategy, 13, 173 19, 23, 30, 32, 55, 57–58, 133,
candidate-centered campaign, 169, 348–349
185–186 Committee for National Health
Cannon, Clarence, (representative), Insurance, 317
58, 98, 229, 231–232, 235, 240 Congress roll call voting behavior
Capitol Hill, 116, 228, 230, 10, 13, 39–40 175–177; travel
233–234, 238, 295 182–184
Carl Albert Center, 54, 63, 95 Congress of Industrial
Carter, Jimmy (president), 7, 42, Organizations (CIO), 317
44, 46–48, 57, 177, 200, 211, Congress of Racial Equality
247, 303, 344, 360 (CORE) 318 [172–173, 309, 318
case files, 21, 132, 135, 138–140 326}
CBS Morning News, 65
CBS, 61, 65, 244, 258
Congressional Biographical 257–260, 263–265, 267,

Directory, 4 269–270, 276, 278–279, 286,
Congressional communications, 290, 292–293, 306–307, 313,
61–66, 92, 177, 186 189– 320, 331, 333–334, 339, 342,
190; congressional frank 171; 347, 351, 353, 361
constituent letters 16, 114, 125, data: code, 10, 189–190, 247–248
136–137, 139–140, 176, 177, 263; collection method 38–39,
190 250–251; press releases 13– 71, 73, 85 260–263; generation
14, 23, 145, 150, 171, 189–190, of 16, 36, 69, 71, 81, 86, 91, 220
319 247–248; qualitative, 63–64, 69,
Congressional Government, 38 78–81, 88, 91, 189–190, 248
congressional oversight, 228–229, 250; quantitative, 36–37, 51–52,
231, 234–235, 241 69, 82, 92, 165–166, 187–189,
Congressional Papers Roundtable 233–235, 252 260; ‘sanitization’
of the Society of American 16–19
Archivists, 8, 56, 144, 154, 352, dataheads, 37, 52, 57, 350
363 DeConcini, Dennis, (senator), 17,
congressional papers, 8, 13, 22, 46, 31, 114, 116, 132–133, 148, 151,
50, 56, 109, 113, 116, 125, 132, 154, 183, 335, 352, 363
144, 153–155, 179, 228, 238, deed of gift, 117–118, 128,
327, 349, 352, 363 131–133, 151, 221, 352
Congressional Staff Directory, 62 Democrat, 175, 182
constituent service, 114, 125, Democratic Message Board, 78, 96
136–138, 178, 190 Democratic National Committee,
container list, 140–141 213, 304
copyright, 127, 129–130, 143, 153, Department of Defense (DOD), 17,
155, 353 109, 212, 321
corruption, 277, 285 Department of Health, New York,
CQ Almanac, 321 284, 292
Crossley, Archibald, 243, 252 Department of Health, Pittsburgh,
284, 292
Daschle, Tom, 19, 31 Department of Public Safety, 285
data, 1, 5–8, 10–11, 13–16, 19–21, Department of Public Works, New
23–26, 30, 36–42, 46–47, 49–52, Orleans, 284–286, 288, 295
54–57, 59, 63, 69–74, 76–77, Department of Public Works,
80–82, 85–86, 90, 92–93, 97, Pittsburgh, 284–286, 288, 295
99, 102, 105, 107, 109–110, Department of the Treasury, 17–18
113, 116, 126, 136, 139, DeWine, Mike (Lt. Governor of
142–145, 147, 150, 154–155, Ohio), 181
162, 165, 169, 174–175, 181, diachronic versus synchronic
183, 187–189, 198, 200–201, analysis, 71, 78, 81–82, 98
203–204, 213–214, 219–220, Dickey, John S., 259–260
222–224, 246–251, 253–254,
Index 363
Dirksen Congressional Center, 52, Forbes, 244, 271

54, 198, 200 Ford Library, 209, 219
Dirksen, Everett, 9, 28–29, 52, 54, Ford, Gerald, (representative), 54,
95, 164, 186, 198, 200, 302–303, 62, 64–65, 165, 209, 214, 219,
352 231, 247, 311, 325, 352
Domenici, Pete, (senator), 117, 156, Franklin D. Roosevelt Library, 255,
165, 172, 189, 196 266
Dulles, Allen W., 231, 233, 235, Freedom of Information Act
238 (FOIA), 215–217, 221, 225, 348
Dulles, John Foster, 231, 233, 235, Freeman, Elsie T., 115, 127, 154
238 Fried, Amy, 88, 243–244, 248, 250,
DW-NOMINATE, 175, 197 267–270, 303, 359
Frisch, Scott, 1, 6–7, 12, 15, 24, 26,
Eagleton, Tom, 16 28, 32, 35–37, 42, 52, 54–56, 72,
Economic Policy Institute, 313 76, 92, 101, 106, 109, 112, 133,
Eisenhower, Dwight D., 178, 198, 147, 151, 154, 171, 196, 200,
208, 213, 230, 232–233, 235, 332–333, 348–349, 359–361
238, 242, 251, 258 front matter, 127, 346
Eisinger, Robert, 248–250, 270 fund raising, 117–118, 163, 186
embarrassing materials, 18
endogenous versus exogenous Gallup, George, 8–9, 28, 118,
determination of preferences, 138–139, 142, 145, 225,
78–79 336–340, 352
Enduring research value, 115 garbage, 275–278, 284–286, 291,
Evans, John, 7, 15, 52–53, 56, 202 297, 362
Evans, Larry, 7, 15, 52–53, 56, 202 George Meany Memorial Archives,
317, 324
Face the Nation, 62, 65 Gerring, John, 71, 84, 86–87, 96,
farm groups, 308 98
federal depository libraries, 107 Get-out-the-vote (GOTV), 182–184
Federal Digital System (Fdsys), 108 Gingrich, Newt, 12, 63, 91
Federal Election Commission, 39 Glenn, John, (senator), 57, 99,
Fenno, Richard, 6, 41, 48, 91, 95, 179–181, 196, 201
98, 170–173, 178–179, 189, 200, Goldwater, Barry (senator),
230, 241 109, 114, 116, 122–124, 129,
filibuster, 301–302, 304, 315 132, 134, 137–141, 148–149,
Finding Aids, 3, 126–127, 140, 175–176, 178, 180–184,
142, 144, 193, 208, 215, 219, 196–198, 200–201, 363
257, 261–262, 321, 339, 347 Google Books, 282, 284, 297
Fiorina, Morris, 74, 76, 97–98 Google Doc, 284–285
Fleischer, Ari, 12, 31 Gov Docs, 102, 106–108, 155
Foley, Tom, 15, 54, 63, 95–96 Government Information Online,
Foote, Joe S., 62, 64–65, 96 108
Government Printing Office (GPO), 263, 267–268, 271, 278–279,

108–109, 130, 153, 242, 297 282–283, 292–293, 295, 300,
graft, 285 303, 306–307, 309–311, 313,
grants, 9, 40, 107, 137, 145, 147, 353, 363
268, 352 historical institutionalism, 68, 70,
Gravel, Mike (senator), 190, 198, 72, 245
201 Home Styles, 171, 200
Greene, Mark, 18, 21, 104–105, Hoover Institution, 131, 155
111, 124, 147, 154 Hoover, Herbert, 10, 131, 155,
Greenstone, David, 300, 326 205–206, 212, 248
House, 4, 12–13, 19, 21, 30–32, 35,
H. Con. Res. 114 41–42, 44, 51, 55–57, 61–66,
Harris, Douglas, 1, 15, 26, 32, 42, 78, 89–90, 92, 95, 98–99, 115,
55, 57, 61, 63, 95, 99, 109–110, 119, 132, 135, 178, 193, 195,
112, 155, 196, 201, 248–250, 198, 200, 202–204, 208–214,
269–271, 307–308, 327, 360 222–223, 225, 229–231, 25, 240,
Harry S. Truman Library, 163, 208, 247, 251, 255–257, 260, 268,
212, 230 271–272, 299, 302–304, 326,
Hartsook, Herb, 17 342, 359–360
Hayden, Carl, (senator), 123, 148, House of Representatives, 30, 32,
231, 240, 361 42, 57, 115, 135, 200, 255–256,
health care policy, 300 268, 360
Health Insurance Association of Huddleston, Walter, (senator), 178
America (HIAA), 312, 325 humanism, 68
health insurance, 299, 311, 317,
325 Idaho Historical Society, 7
Heinz History Center, 283–284 Igo, Sarah, 244, 246, 251, 266–267,
Heith, Diane, 247–250, 271 271
historians, 1, 26, 31, 38, 71–72, immigration policy, 307–308, 311
84–86, 97, 101, 105, 110, indeterminacy versus determinacy,
115–116, 138, 150, 153, 156, 78
175, 256, 263, 280, 282, 286, Information Task Force, 62–64, 91
300, 346 institutional change, 63, 65–67, 69,
Historic New Orleans Collection, 71, 73, 75, 77, 79, 81–83, 85, 87,
284 89, 91, 93–95, 97, 99
historical, 7, 15, 19, 26, 31, 51, institutional change: demand and
61, 63, 67–73, 77, 79, 81–86, supply sides, 66–67
88–89, 91–93, 95, 97, 99–100, interest group, 208, 305–311, 315,
104, 114–115, 123, 133–134, 320, 325–326
137, 148–149, 153–154, 180, International Association of
195–199, 201–202, 204–205, Machinists (IAM), 318
220, 223–224, 226, 240, International Longshoremen, 317
245, 249–251, 254, 256–257,
Index 365
International Workers of the World 210, 212, 220, 224, 232, 246,
(IWW), 317 249–251, 267, 289, 295, 315,
Interstate Commerce Act of 1887, 333, 335, 348
74 Lexis-Nexis, 14, 318
Interuniversity Consortium for Library of Congress (LOC), 4, 95,
Political and Social Research, 8 205–206, 225, 255, 318
interview, 7, 15, 18, 143, 149, 183, Literary Digest, 244, 252–253, 270,
197–198, 249, 260, 268, 303, 272
313, 324, 335, 348 Livingston, Bob, 12, 31
interviews, methods of 14–15, 17, Lowi, Theodore, 35, 58
19–20, 24–25, 41, 213–214, 249, Lucas, Scott (senator), 129, 155,
306–307, 310–311 186
Jacoby, Sanford, 78, 81, 99
Mackaman, Frank, 29, 52, 54
Johnson, Lyndon, 54–55, 57–58,
Mahon, George, 232, 240
99, 219, 227, 230, 241, 251,
Majority Leader, 302
301–303, 324, 359, 362
Managing Congressional
Journal of the American Medical
Collections, 119, 155, 363
Association (JAMA), 320
Mansfield, Mike, 302
Keating Five, 181 Matsunaga, Masayuki, (senator),
Keiser, J. Roddy, 62 140
Kelly, Sean, 1, 6–7, 12, 15, 24, Mattingly, Mack, 177
26–28, 32, 35–37, 42, 47, 52, Mayhew, David, 39, 58, 163, 168
54–56, 58, 72, 76, 92, 101, McCloskey, Paul N. ,
106, 109, 112, 133, 137, 147, (Congressman), 131, 155
151, 154, 171, 196, 200–201, McConnell, Mitch (senator), 178
331–333, 335, 347–349, 360 McCormack, John, 32, 57, 62–63,
Kennedy, John F., 208–210, 228, 65–66, 95, 360
230, 238, 240–241, 251, 271, McDade, Joseph, 51
359 McKinley, William, 95, 208, 308
Kerry, John, 24, 31 McWhite, Leigh, 21
Key, Valdimer O., 26, 37, 41 Meany, George, 302
Knights of Labor, 310 media, 14–15, 61–62, 64–66, 89,
112, 117, 125–126, 129, 134,
labor movement, 299, 301–303, 162, 170, 185, 189–190, 193,
313, 325 196–197, 210, 235, 244, 248,
Lazarsfeld, Paul, 258, 266–267, 251, 255, 257–258, 262, 273,
270–271 313, 322, 360
Leadership Conference on Civil Meet the Press, 62
Rights, 318 methodological individualism, 37
League of Women Voters, 318 Michel, Robert, (R–IL), 28, 52, 90,
letters, 13, 28–29, 47, 64, 89–90, 95, 270
103, 136, 138, 177, 191, 208,
money, 8–9, 47, 118, 124, 162, 186, National Organization of Women
202, 228, 232, 256, 322, 332, (NOW), 318
335–338, 346, 352, 361 National Resources Defense
Montoya, Joseph, 172 Council (NRDC),
“more product, less national security–classified records,
process” (MPLP), 124–125, 147 access, procedures, 217, 230,
Morris, Thomas Gayle, 232, 237
(Congressman), 74, 95, 97–98, National Small Business
132, 155, 241, 352 Association (NSBA), 309, 320
Municipal Sanitation in the United National Urban League, 318
States, 279, 297 NBC, 61
municipal records, 282–283, 286, New Deal, 301
291 new institutionalism, 68, 70, 75, 77,
municipalities, 275, 279, 281, 362 82, 99–100
Muskie, Edmund, (senator), 139 New Mexico State University
Libraries, 132
National Abortion Rights Action New Orleans, 275–276, 280–284,
League (NARAL), 318 286–289, 295, 297
National Aeronautics and Space New York Times Magazine, 65
Administration (NASA), 179 New York Times, 65, 165, 168, 186,
National Archives, 36, 70, 99, 200, 266
137, 206–207, 216–217, 221, Newspaper Guild, 317
230–233, 237–238, 255, 257, Nixon, Richard, 62, 64, 207, 219,
262, 266, 268, 300, 320, 343, 235, 247, 251, 271
359 Northwest Digital Archives, 4, 31
National Archives and Records
Administration (NARA), 137, O’Neill, Tip, 63, 95, 249
206, 217, 320 Office of National Drug Control
National Association for the Policy (ONDCP), 17
Advancement of Colored People ordinances, 276, 283, 291–292
(NAACP), 317 organized labor, 299–301, 303,
National Association of 310–312, 315, 317, 325, 361
Manufacturers (NAM), 308, 318 Orren, Karen, 70, 84, 96, 100, 245,
National Consumers’ League, 318 272
National Council of La Raza Oval Office, 205, 207
(NCLR), 320
National Federation of Independent Panama Canal Treaties, 177
Business (NFIB), 309 Parker, David, 1, 7–8, 16, 22, 42,
National Journal, 65, 200 109, 125, 150–151, 161, 169,
National Labor Relations Act, 315 171, 186, 198, 201–202, 332,
National Opinion Research Center, 361
260 participant-observation, 41
party-centered campaign, 164
Index 367
pension, 311, 320 presidential library, 213–214, 219,

personal papers, 113, 130, 144, 176, 264
214, 285, 291, 343 Presidential Recording and
photocopying, 41, 218, 236, 322, Materials Preservation Act of
337, 344–345, 348 1974, 206
Photoshop, 291 Presidential Records Act of 1978,
Picasa, 291, 345 206, 221
Pierce, Walter M., 256–257, 266, press release, 150, 188–189
268 primary resources, 4, 103, 249–250
Pittsburgh, 241, 276–277, 280–281, primary sources, 4, 101, 103, 106,
283–285, 295 144, 332
Planned Parenthood, 176, 197 Professional Air Traffic Controllers
Political Archives Reader, The, , 57, Organization (PATCO), 318
99, 154–155, 360 Progressive Alliance, 313
Political Graveyard, 256 Progressive Era, 279, 307–308
political machines, 280 Proquest’s Archive Finder, 4
Political Monopolies in American provenance, 3, 103, 125, 127–128,
Cities, 292, 297 211
political monopolies, 292, 297 public opinion, 27, 47, 57, 69, 77,
political scientists, training of 1– 88–89, 92, 99, 176, 203–204,
2, 35–36, 40, 67–69, 71, 75–76, 210, 213–214, 220, 222,
87–89, 93, 96 225–226, 244–261, 263, 265,
politicians, 2, 6, 13, 17, 20, 26, 267, 269–273, 306, 315, 326,
46–48, 66, 80–81, 172–174, 250, 359, 362
271, 300, 314–315, 319, 347 public opinion and the American
polling, 7, 13, 136, 150, 177, state, 235, 250
188, 213–214, 219, 222, 225, public opinion networks, 245
243–251, 256, 258–260, 262, Public Opinion Quarterly, 253,
264, 266–267, 270–272, 320, 256, 270–272
359 public opinion research, historical
polls, 7, 22, 39, 92, 126, 150, turn 39, 247–252
170, 177, 182, 187–189, 193, public Speakership, 61–63, 66, 99,
204, 244, 247–250, 252–253, 271
255–257, 259–260, 264, 267,
269–272, 306, 333 qualitative, 5, 11, 15, 23, 69–71, 76,
post-participant observation, 96, 143–144, 224, 226, 246, 248,
173–174 266, 271, 333
presidential archive, 213–214, 219, quantitative, 1, 5, 11, 15, 40, 51–52,
264 54, 56, 69–71, 73, 76, 82, 92–93,
Presidential Library Acts of 1955 96, 109, 143, 190, 194–195, 203,
and 1986, 225 234–235, 244, 246–247, 249,
251–252, 263, 276, 278, 293,
305, 313, 333, 335
Rayburn, Sam, 62–63, 65–66, 98 restrictions, 3, 18, 117, 127,

Reagan, Ronald, 165–166, 182, 131–133, 149, 151–152,
188, 247, 344 191–192, 217, 221, 347
Report on the Social Statistics of retention schedules, 115–116
Cities, 279, 297 Richard B. Russell Library, 202,
repository fees, 337 239–240
repository, 1, 3, 8–9, 16–18, Rockefeller Archive Center (RAC),
21, 23, 103–104, 106, 111, 258, 266, 268
116–119, 125–130, 136, 139, Roof, Tracy, 83, 88, 100, 299, 301,
143, 147–148, 150, 152, 155, 327, 361
268, 283, 333–337, 339–340, Roosevelt, Franklin D., 208, 212,
343–345, 347, 350, 352 244, 252, 255, 259, 266
Republican, 52, 63, 78, 81, 90, 96, Roper Center’s Public Opinion
164–165, 177, 181–184, 186, Archives, 260
189, 213, 300–302, 333 Roper, Elmo, 243–246, 252, 260
Republican Committee on Ross, Dorothy, 68, 83, 87, 100
Committees, 333 Rudd, Eldon, (Congressman),
Republican party, 52, 181, 183 134–135, 156
Republican Theme Team, 78 Rules Committee, 302, 304
research, 1–16, 18, 20, 22, 24–28, Russell, Richard, (senator),
30, 32, 35–59, 61–64, 66–84, 86, 197, 202, 227–229, 232, 235,
88–102, 104–110, 112, 114–116, 239–241, 258, 349
118, 120–122, 124, 126, 128,
130, 132–136, 138–144, San Francisco, 259, 280–281, 283
146–148, 150–156, 162–164, Sanitarians, 279
166–170, 172, 174–176, SAS, 69
178, 180, 182, 184, 186, 188, Schattschneider, E.E., 79, 100
190–198, 200–202, 204, 206, Schmitt, Harrison “Jack”, (senator),
208, 210, 212–216, 218–230, 114, 150–151, 164–168, 172,
232–234, 236–240, 242–246, 174, 188, 193, 196–198, 202
248–256, 258, 260–264, Schulz, William (Senate candidate),
266–273, 276, 278, 280–286, 175–176, 181, 183–184,
288–292, 294–296, 299–318, 197–199, 202
320–324, 326, 331–342, 344, scientism, 68
346–350, 352, 359–363 search strategies, 141, 334
research, budget 336–338, 352 Second face of power, 19, 174
356; cities,275–276 278–281; selection bias, 99, 194, 201, 220
funding, 268, 338–340 352;
grants, 9, 41, 54, 268, 339 352;
qualitative, 246, 249 266; travel
plans, 261–262, 331–335, 357
Index 369
Senate, 4, 7, 16, 19, 31, 55–57, Steiger, Sam (Senate candidate),

63, 98, 100, 114–115, 132, 150, 183–184
156, 161–164, 168, 170–172, Stevens, Ted, (senator), 118, 120,
175, 177–180, 184, 186–188, 156
195, 197–202, 229–232, 235, Supreme Court, 55, 58, 132, 302
241–242, 255–256, 268, 299, surveys, 10, 24, 26, 39–40, 46–47,
301–302, 304, 315, 335, 349, 106, 115, 177, 181–182, 187,
362 203, 246, 252–253, 267, 271,
Senate Historical Office, 31, 256 279, 305, 313
Senate office handbook, 114 Symington, Stuart, (senator), 233
series, 3, 15, 17–18, 21, 51, Symms, Steve, 7
65, 114, 125, 129, 134–138,
140–141, 155–156, 163, 180, Taft-Hartley Act 14(b), 301, 315
182, 187, 197–198, 245, 250, Talmadge, Herman, (senator),
267, 306, 341, 343–344, 346 176–177, 197, 202
Service Employees International taxpayer, 113
Union (SEIU), 317 Truman, Harry S., 163, 208,
Shadegg, Steven, 182–184, 198, 212, 230, 234, 241, 243–244,
202 258–259, 266, 271–272, 301,
Sierra Club, 309 303, 315–316, 324, 359
Skemer, D.C., 58, 67–70, 88, 96, twenty-one-day rule, 302
100–101, 105, 107, 113, 116, U.S. News and World Report, 65
156 uncertainty, 76–77, 79
Skowronek, Stephen, 70, 84, 87, union, 88, 301, 310, 313, 317–318
96–97, 100, 245, 272 United Auto Workers (UAW), 303,
Smathers, George (Library), 18 317
Social Security Administration, United Farm Workers (UFW), 317
311, 320 United Garment Workers of
Social Security Death Index, 143 America (UGWA), 318
Society of American Archivists United Kingdom, 4
(SAA), 9, 103, 143, 153–156, United States, 4, 31, 105, 131, 135,
318, 320, 352, 363 142, 153, 156, 163, 229, 233,
Southern Christian Leadership 245, 255, 270, 279, 297, 299,
Conference (SCLC), 318 311, 318, 325–327, 361–362
Sparrow, Bartholomew H., 101 University of Chicago Library
Speakers, 61–62, 64, 117, 164, 174, Special Collections and
186 Research Center, 266
SPSS, 69 University of Hawaii, 140
standard of living, 300 University of Pittsburgh archives,
Stanton, Frank, 258, 267, 273 283
Stata, 69 University Publications of America
(UPA), 318
veterans’ groups, 310 White House, 41–42, 44, 89, 99,

veto strategy, 48 203–204, 208–214, 222–223,
Vinson, Carl, (representative), 225, 247, 251, 271–272, 303,
230–231, 241 359
White House Office of Records
Walter P. Reuther Library (Wayne Management (WHORM), 209,
State University), 317 225
Washington Star, 65 Williams Research Center, 283–284
Welker, Frank, (senator), 164, 174, Wilson, Woodrow, 38
186 women’s groups, 308, 310, 318
Whip counts, 15, 45, 47 World War II, 300–301, 307–308,
310–311
WorldCat, 142–143, 284
Wright, Jim, 63, 95
About�the�Contributors
David M. Barrett is a professor of political science at Villanova

University and holds a PhD from the University of Notre Dame. He
specializes in the U.S. presidency and U.S. foreign policy. Professor
Barrett’s most recent book, The CIA and Congress: The Untold Story
from Truman to Kennedy (Lawrence: University Press of Kansas), is based
on research that he conducted at National Archives facilities as well as
at archives around the country holding the papers of sixteen deceased
members of Congress. He also has published two other books on President
Lyndon Johnson and the Vietnam War. He is currently working on two
book projects, one about intelligence aspects of the Cuban missile crisis,
the other a broader history of the CIA, Congress, and the Kennedy White
House.
Amy Fried is a professor of political science at the University of Maine

and holds a PhD from the University of Minnesota. For Pathways to
Polling: Crisis, Cooperation and the Making of Public Opinion Profes-
sions (New York: Routledge), she worked with an array of archival mate-
rials pertaining to important figures in the polling industry as well as
government, foundation, and business enterprises that were involved in
the early years of survey and market research. She is also the author
of Muffled Echoes: Oliver North and the Politics of Public Opinion
(New York: Columbia University Press). Professor’s Fried’s current
research builds on Pathways to Polling and focuses on the incorpora-
tion of race in U.S. government opinion research. She blogs at http://
www.pollways.com.
Scott A. Frisch is a professor and chair of political science at Cali-

fornia State University Channel Islands. He received his PhD from Clare-
mont Graduate University. Professor Frisch is the author of The Poli-
tics of Pork: A Study of Congressional Appropriations Earmarks and a

coauthor with Sean Q Kelly of Committee Assignment Politics in the
U.S. House of Representatives (Norman: University of Oklahoma Press),
Jimmy Carter and the Water Wars: Presidential Influence and the Poli-
tics of Pork (Amherst, NY: Cambria Press), and Cheese Factories on the
Moon: Why Earmarks Are Good for American Democracy (Boulder, CO:
Paradigm Publishers).
Douglas B. Harris is an associate professor in the Department of

Political Science at Loyola University, Maryland, and holds a PhD from
Johns Hopkins University. His research on Congress, political parties,
and media politics includes articles in Political Science Quarterly, Polit-
ical Research Quarterly, P.S.: Political Science & Politics, The New
England Journal of Political Science, and Presidential Studies Quarterly.
His archival research in the papers of numerous congressional leaders
(including every Speaker of the House from John McCormack to Newt
Gingrich) is featured in articles in Legislative Studies Quarterly, Amer-
ican Politics Research, The Historian, and Congress & the Presidency as
well as in chapters in Winning with Words: The Origins & Impact of Polit-
ical Framing (New York: Routledge) and An American Political Archives
Reader (Lanham, MD: Scarecrow Press). Along with three colleagues,
Dr. Harris is a coauthor of The Austin-Boston Connection: Fifty Years of
House Democratic Leadership (College Station: Texas A&M University
Press), which made use of dozens of archival collections spanning twen-
tieth-century congressional history.
Sean Q Kelly is a professor of political science at California State

University Channel Islands. He received his PhD from the University of
Colorado. Professor Kelly is a former American Political Science Associ-
ation congressional fellow (1993–1994) and is a coauthor of three books
with Scott Frisch, most recently Cheese Factories on the Moon: Why
Earmarks are Good for American Democracy (Boulder, CO: Paradigm
Publishers) and Jimmy Carter and the Water Wars: Presidential Influence
and the Politics of Pork (Amherst, NY: Cambria Press). He is currently
About the Contributors 373
working with Scott Frisch on a book, based on archival data, about the
politics of congressional appropriations earmarks.
Michael Lotstein is an arrangement and description archivist at Yale
University’s Sterling Memorial Library and holds an MA from Arizona
State University. He has been a practicing archivist since 2006 and has
worked on a variety of political collections, including the papers of
Senator Carl T. Hayden, Congressman John J. Rhodes, and Congressman
Bob Stump, each of which documents the growth and development of
the western United States in the twentieth century. He is currently over-
seeing the processing of manuscript collections related to the history of
Yale University and greater Connecticut.
David C. W. Parker is an assistant professor of political science at

Montana State University–Bozeman and holds a PhD from the Univer-
sity of Wisconsin–Madison. He utilized archival materials in his book
The Power of Money in Congressional Campaigns, 1880–2006 (Norman:
University of Oklahoma Press) to debunk the notion that party-centered
campaigns disappeared from American politics in the mid-twentieth
century. Dr. Parker has published articles in Legislative Studies Quar-
terly and Congress and the Presidency on congressional investigations and
the use of congressional office resources to create particular constituent
impressions. In addition to his continuing research in these areas, he is
currently writing a book on senators who lose reelection using—what else
—archival materials.
Tracy Roof is an associate professor of political science at the Univer-

sity of Richmond. She received her PhD from Johns Hopkins University.
Her research examines the influence of organized labor in the American
legislative process and the role of institutions in shaping American public
policies. She used archival research to explore labor’s lobbying on full
employment policy, health care reform, labor law reform, and the reform
of congressional procedures. Dr. Roof is the author of American Labor,
Congress, and the Welfare State, 1935–2010 (Baltimore: Johns Hopkins
University Press, 2011) as well as various articles on organized labor’s
political activities. She served as an American Political Science Associa-

tion congressional fellow in the Senate.
Brandon Rottinghaus is an associate professor of political science at
the University of Houston and holds a PhD from Northwestern Univer-
sity. His research interests include the presidency, public opinion, execu-
tive-legislative relations, and research methods; his work on these subjects
has appeared in several academic journals. Dr. Rottinghaus is the author
of The Provisional Pulpit: Modern Conditional Presidential Leadership
of Public Opinion (College Station: Texas A&M University Press). He
is the founding designer of the Presidential Proclamations Project at the
University of Houston, an online resource documenting the presidential
use of unilateral powers through executive proclamation.
Patricia Strach is an associate professor in the Departments of Polit-

ical Science and Public Administration and Policy at the University
at Albany, State University of New York. She holds a PhD from the
University of Wisconsin–Madison. Her research examines the relation-
ship between social and political institutions in American public policy.
She is currently using archival research in her projects on why people
turned to government for public policy solutions in late twentieth-century
disease advocacy and in nineteenth-century municipal garbage collection.
Dr. Strach is the author of All in the Family: The Private Roots of American
Public Policy (Stanford, CA: Stanford University Press) as well as arti-
cles appearing in Political Research Quarterly, Journal of Policy History,
Polity, and American Politics Research. From 2008 to 2010, she was a
Robert Wood Johnson scholar in health policy research at Harvard Univer-
sity.
Kathleen Sullivan is an associate professor in the Department of Polit-

ical Science at Ohio University in Athens, Ohio. She holds a PhD from the
University of Texas at Austin. Her research examines political authority
and civic formation in nineteenth-century municipalities in the United
States in studies ranging from garbage collection (with Patricia Strach) to
juvenile courts (with Carol Nackenoff) to the regulation of sailors in port
About the Contributors 375
cities. Dr. Sullivan is the author of Constitutional Context: Women and

Rights Discourse in 19th-Century America (Baltimore: Johns Hopkins
University Press) and has published articles in Studies in American Polit-
ical Development and Political Research Quarterly.
Linda A. Whitaker is a certified archivist and librarian at the Arizona

Historical Foundation in Tempe, Arizona. She specializes in congressional
papers and has processed or managed the collections of Senators Barry
Goldwater, Dennis DeConcini, and Paul Fannin. She was a member of
the Editorial Advisory Board for the Managing Congressional Collections
manual. She is past chair of the Congressional Papers Roundtable Steering
Committee and is a current member of the Appraisal and Acquisitions
Section Steering Committee. She also serves on the Society of Amer-
ican Archivists’ Reappraisal and De-accession Development and Review
Team, which is charged with developing national guidelines for removing
or transferring collections. She is a founding member of the Arizona
Archives Summit, a federally funded statewide initiative for collaborative
collections management.
View publication stats

Dar Ps Manuscript

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dar Ps Manuscript

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Doing Archival Research in Political Science

Book · January 2012

Scott A. Frisch Sean Q Kelly

SEE PROFILE SEE PROFILE

Doing Archival Research View project

Jimmy Carter and the Water Wars View project

The user has requested enhancement of the downloaded file.

Scott A. Frisch, Douglas B. Harris,

All rights reserved

No part of this publication may be reproduced, stored in or introduced

Requests for permission should be directed to:

Library of Congress Cataloging-in-Publication Data

Doing archival research in political

List of Figures ................................................................ vii

Chapter 9: Digging Through Trash

Figure 1: Page 1 of a list of members that the Carter White

Figure 2: Whip counts from the papers of Representative Tom

Figure 3: Archivist Mary Anne Hamblen in the storage warehouse

Figure 4: Personal and political papers of Senator Barry M.

Figure 5: Personal and political papers of Senator Barry M.

Figure 6: Payroll, Department of Public Works, 1883 ................. 288

Table 1: The data available in archival collections. ...................... 49

Table 2: Ready and latent data in congressional papers

Table 3: Container list from the 1964 presidential campaign series

Table 5: Senator Barry Goldwater and going home, 1975–

Table 6: Comparison of Republican vote percentages in

Table 7: Change in voter turnout between 1976 and 1980 in

Table 8: Library of Congress presidential manuscript

Table 9: NARA presidential library websites and locations ........... 207

Table 10: Budget worksheet ............................................... 355

Table 11: Travel checklist ................................................. 357

Scott A. Frisch, Douglas B. Harris,

In the discipline of history, archival research is common. In political

This book is aimed at political scientists. We simultaneously intend to

opportunity to explore different—and, we argue, more innovative and

In conceiving this book, we hope to produce a volume that we wish had

Once a collection is processed, it is made available to researchers.

Reason One: This May Not Generate Publishable Findings

congressional scholars have long used committee assignments as an indi-

Reason Two: The Information Sought May Not Be Found

Remember a piece of advice that nearly every author in this volume

Reason Three: A Great Deal of Time Is Needed

In short, researchers have the time; it is just a matter of how they

Archival travel, actually, is a fantastically efficient use of a researcher’s

Reason Four: Money Is Needed to Do This Research

of the Society of American Archivists keeps a list of these funds at

Reason Five: Colleagues Say It Is a Waste of Time

use of archival sources? If it is the former, a return to the drawing board

If colleagues have reservations about a research design, listen very care-

Reason One: It Will Improve Research

Reason Two: It Will Inform Teaching

The following paragraph from Fleisher’s memo to Livingston is a gem

The way to get a message out is to constantly repeat the mantra.

• Education, locally controlled

Literally, this is an eight-second sound bite. I recommend you

Other examples of documents found in congressional papers that can

Reason Three: New and Better Data Will Be Introduced

Reason Four: Faulty and Selective Memories Are Less Common

The archival record, in contrast, is often not subject to this problem.

Reason Five: Other Subjects of Research Will Be Uncovered

Senator Tom Eagleton’s papers included a weekly tally of constituent

Reason One: Political Papers Are Sanitized

Archivists work closely with potential donors before papers arrive at